@sema-agent/client-core 0.94.0 → 0.95.0

package/README.md

@@ -35,7 +35,7 @@ Renamed from **`@sema-agent/wire-cc-adapter`** (0.1.x, deprecated — see *Migra
 
 ## Scope
 
-**Version:** 0.94.0
+**Version:** 0.95.0
 
 - **Today** — the adapter seam, the whole `adapt()` pipeline (all 14 A-layer arms plus the
   B/D/E tool-card layers), the notification/caps/model families, the adapter kernel (stream driver
@@ -67,12 +67,12 @@ Renamed from **`@sema-agent/wire-cc-adapter`** (0.1.x, deprecated — see *Migra
     against — the tables live upstream precisely so this package does not keep a second copy that can
     fall behind. The browser bundle really bundles the SDK through (the portability guard would
     exit 3 rather than quietly mark it external).
-  - The declared floor is `>=18.1.0` (raised from `>=12.0.1` in 0.94.0: the request-side permission-mode vocabulary, the decision-attribution key type, the cancellation-cause list, the status-derived coarse-code list and the routing-failure denial word are now taken from the SDK itself, all of which it declares from 17.x or 18.x on; before that raised from `>=11.3.0` in 0.84.0: the per-session background listing (`sessions.background`), its capability bit `capabilities.background.listFace` and the closed background-status vocabulary are typed from 12.x on, and the package reads all three, so it no longer compiles against 11.x; before that raised from `>=11.2.1` in 0.80.0: a deny decision may now name its settler — `settledBy: "policy"` is typed on both the durable decide body and the live respond body from 11.3.0 on, and the gate record's settlement vocabulary carries its thirteenth word `policy_refused`, which this package reads to place a refusal in the policy bucket rather than the person's; the package now compiles against 11.3.0 and no consumer ships 11.2.x any more, so the older floor lost its witness; before that raised from `>=11.0.1` in 0.79.0: `capabilities.mcpProbe`, the MCP probe face (`mcpCapabilities` / `probeMcp` with `McpProbeFace`) and the write receipt's third liveness arm (`stillLive: "unknown"`) are typed from 11.2.x on, 11.2.1 adds `TaskRequest.approverPosture` and the `mandated` approval-frame key to the types and the runtime key anchor, the package now compiles against 11.2.1, and no consumer ships 11.0.x / 11.1.x any more, so the older floor lost its witness; before that raised from `>=9.8.1` in 0.78.0: `DeniedBy` carries its tenth word `read_boundary`, `rules.write` answers a `stillLive`-discriminated body, `RemovalLiveness` / `RuleWriteRequest` / `RuleWriteResult` / `RuleWriteBehavior` are exported from the SDK root and `TaskRequest.excludeAllTools` is typed from 11.x on, the package now compiles against 11.0.1, and no consumer ships 9.8.x any more, so the older floor lost its witness; before that raised from `>=9.7.1` in 0.75.0: `Capabilities.deviceExecutor.management` is typed from 9.8.x on, the package now compiles against 9.8.1, and no consumer ships 9.7.x any more, so the older floor lost its witness; before that raised from `>=9.6.0` in 0.74.0: `Capabilities.approvalsStreamLive` / `.executionLane`, `LivePendingRow.frame`, the `live_*` approval-stream events and `gates[].toolCallId` are typed from 9.7.x on, and no consumer ships 9.6.0 any more, so the older floor lost its witness; before that raised from `>=9.4.0` in 0.71.0: the `tool_disclosure` / `tool_progress` frames and `ToolApprovalFrame.readRootCandidate` are typed there; earlier: raised from `>=8.8.0` in 0.69.0: the `reasoning_end` frame and `McpStatusPanel.lastLegMcp` are typed from 9.4.0 on), and it is *witnessed*: the guard checks that an actually
+  - The declared floor is `>=19.0.0` (raised from `>=18.1.0` in 0.95.0 with the sdk 19 pick-up: the ask-row ending-cause table, the run-death verdict codes and the terminal-frame rule now come from the sdk's own tables; before that raised from `>=12.0.1` in 0.94.0: the request-side permission-mode vocabulary, the decision-attribution key type, the cancellation-cause list, the status-derived coarse-code list and the routing-failure denial word are now taken from the SDK itself, all of which it declares from 17.x or 18.x on; before that raised from `>=11.3.0` in 0.84.0: the per-session background listing (`sessions.background`), its capability bit `capabilities.background.listFace` and the closed background-status vocabulary are typed from 12.x on, and the package reads all three, so it no longer compiles against 11.x; before that raised from `>=11.2.1` in 0.80.0: a deny decision may now name its settler — `settledBy: "policy"` is typed on both the durable decide body and the live respond body from 11.3.0 on, and the gate record's settlement vocabulary carries its thirteenth word `policy_refused`, which this package reads to place a refusal in the policy bucket rather than the person's; the package now compiles against 11.3.0 and no consumer ships 11.2.x any more, so the older floor lost its witness; before that raised from `>=11.0.1` in 0.79.0: `capabilities.mcpProbe`, the MCP probe face (`mcpCapabilities` / `probeMcp` with `McpProbeFace`) and the write receipt's third liveness arm (`stillLive: "unknown"`) are typed from 11.2.x on, 11.2.1 adds `TaskRequest.approverPosture` and the `mandated` approval-frame key to the types and the runtime key anchor, the package now compiles against 11.2.1, and no consumer ships 11.0.x / 11.1.x any more, so the older floor lost its witness; before that raised from `>=9.8.1` in 0.78.0: `DeniedBy` carries its tenth word `read_boundary`, `rules.write` answers a `stillLive`-discriminated body, `RemovalLiveness` / `RuleWriteRequest` / `RuleWriteResult` / `RuleWriteBehavior` are exported from the SDK root and `TaskRequest.excludeAllTools` is typed from 11.x on, the package now compiles against 11.0.1, and no consumer ships 9.8.x any more, so the older floor lost its witness; before that raised from `>=9.7.1` in 0.75.0: `Capabilities.deviceExecutor.management` is typed from 9.8.x on, the package now compiles against 9.8.1, and no consumer ships 9.7.x any more, so the older floor lost its witness; before that raised from `>=9.6.0` in 0.74.0: `Capabilities.approvalsStreamLive` / `.executionLane`, `LivePendingRow.frame`, the `live_*` approval-stream events and `gates[].toolCallId` are typed from 9.7.x on, and no consumer ships 9.6.0 any more, so the older floor lost its witness; before that raised from `>=9.4.0` in 0.71.0: the `tool_disclosure` / `tool_progress` frames and `ToolApprovalFrame.readRootCandidate` are typed there; earlier: raised from `>=8.8.0` in 0.69.0: the `reasoning_end` frame and `McpStatusPanel.lastLegMcp` are typed from 9.4.0 on), and it is *witnessed*: the guard checks that an actually
     installed SDK at that line still exports every value-level symbol this package imports and still
     declares `TaskStats.costMicroUsd` (the key `costOrNull` reads). A floor nobody ever ran is a
     promise, not a contract.
-- **Support floor:** the oldest server, engine and SDK this package still supports — server **7.106.0**,
-  engine **7.35.2**, SDK **18.1.0** (SDK raised from 12.0.1 in 0.94.0; server / engine raised from 7.104.0 / 7.33.8 in 0.92.0). The values are
+- **Support floor:** the oldest server, engine and SDK this package still supports — server **7.115.0**,
+  engine **7.40.1**, SDK **19.0.0** (server / engine raised from 7.106.0 / 7.35.2 in 0.95.0 and from 7.104.0 / 7.33.8 in 0.92.0; SDK raised from 18.1.0 in 0.95.0 and from 12.0.1 in 0.94.0). The values are
   declared in `scripts/support-floor.json` and exported at runtime as `SUPPORT_FLOOR`; servers below the floor
   are no longer witnessed, and the readings that existed only for them were retired in the same release.
 - The direction is constitutional: **sdk (wire) + agent-types (vocabulary) → this package**,
@@ -293,7 +293,7 @@ guard still cross-checks the table by name).
 
 | Suite | What it guards |
 |---|---|
-| `scripts/run-client-core-pure-test.mjs` | Consumer-view behaviour of every moved-in module; segment floors (B1–B8 + C) that only move up The non-throwing attachments verdict names a retired key, and gives its reason and way out, from the retirement rule itself; a negative control plants a second retired key in a copy of the build and checks that it is reported by its own name. The configuration refusal registry includes the inline-image refusal code, whose one sentence names no codes or identifiers, recognises only that code, and is checked against both the installed engine (minted in exactly one place) and the receiving server's older engine (not minted at all). The background-agent task-output reader hands the state and error lines that directly follow the status line to optional slots, positionally: lines written inside the agent's own result text never become slots; checked on twelve forms minted by the installed engine's own functions and by the receiving server's older engine. It also hands the result label's parenthetical note and the header bytes not mapped to a named slot to optional slots, with a rebuild invariant (status line, state, error, the unmapped header bytes, the result label and the output rebuild the unwrapped text, up to four named normalizations) checked on those same forms and on a seeded random pool of forged headers. A Bash command that is moved to the background mid-flight (by the user, or by the engine on a timeout or a queued message) is registered for the background panel by the same event and the same record as a model-requested background command: the reader recognises only the structured detached result (never the receipt prose, never the request's background flag), at most one registration is minted per card (a card both readers recognise yields one, taken from the structured result), the package keeps no record of what it has already published (the same task id on a second card is registered again, because only the consumer knows whether it still has the row), the receipt reader and the detached reader both stand down when the structured result says the engine failed to register, and the stream is compared byte for byte against the previous published build on inputs without a detached result (including two receipts for one task id), differing only by that one event on inputs that have one. |
+| `scripts/run-client-core-pure-test.mjs` | Consumer-view behaviour of every moved-in module; segment floors (B1–B8 + C) that only move up The non-throwing attachments verdict names a retired key, and gives its reason and way out, from the retirement rule itself; a negative control plants a second retired key in a copy of the build and checks that it is reported by its own name. The configuration refusal registry includes the inline-image refusal code, whose one sentence names no codes or identifiers, recognises only that code, and is checked against both the installed engine (minted in exactly one place) and the receiving server's older engine (not minted at all). The background-agent task-output reader hands the state and error lines that directly follow the status line to optional slots, positionally: lines written inside the agent's own result text never become slots; checked on twelve forms minted by the installed engine's own functions and by the receiving server's older engine. It also hands the result label's parenthetical note and the header bytes not mapped to a named slot to optional slots, with a rebuild invariant (status line, state, error, the unmapped header bytes, the result label and the output rebuild the unwrapped text, up to four named normalizations) checked on those same forms and on a seeded random pool of forged headers. A Bash command that is moved to the background mid-flight (by the user, or by the engine on a timeout or a queued message) is registered for the background panel by the same event and the same record as a model-requested background command: the reader recognises only the structured detached result (never the receipt prose, never the request's background flag), at most one registration is minted per card (a card both readers recognise yields one, taken from the structured result), the package keeps no record of what it has already published (the same task id on a second card is registered again, because only the consumer knows whether it still has the row), the receipt reader and the detached reader both stand down when the structured result says the engine failed to register, and the stream is compared byte for byte against the previous published build on inputs without a detached result (including two receipts for one task id), differing only by that one event on inputs that have one. The attachments env lane's product default now also carries the todo reminder (one default table shared by the absent, empty, unknown-word and `default` paths); `off` removes the key and an explicit list that leaves it out does not get it. |
 | `scripts/run-client-core-portability-test.mjs` | Kernel / A-layer / index import closures, the runtime-dependency equality gate, barrel reachability, and a real esbuild `--platform=browser` bundle |
 | `scripts/run-client-core-diff-test.mjs` | Differential equivalence against the CLI reference bridge + replay-id invariant + ledger round-trip |
 | `scripts/run-seat-contract-keys-test.mjs` | The seat IPC contract: verb list ↔ SPEC ↔ types, element-wise. Since 0.83.4 the seat approval request also carries the four card bits that say whether an approval must be asked and where it came from (`mandated`, `mandate`, `ruleOffersAbsence` — the parked-row form of mandated, so the seat and the card give the same `approvalIsMandated` answer — and `origin`, which the resident-posture predicate reads), and they reach the seat through `toolPermissionRequestAskBits`, which reads them exactly as the card request does — a strict own `true`, one of the closed mandate words byte for byte, a non-empty own string — so for a host that builds the seat request through that crossing an off-list word, an empty string or an inherited key never lands on the seat; the seat validator does not become a second judge that could drop a safety ask over one bit, which leaves a host that builds the request by hand to apply the same narrow read itself |
@@ -304,22 +304,22 @@ guard still cross-checks the table by name).
 | `scripts/run-sdk-floor-test.mjs` | The SDK version floor — and, more to the point, that the *installed* type declarations still carry the keys this package reads — including, from 0.80.0, the three declarations that justify the floor itself: the key naming who settled a refusal on both decision legs, and the thirteenth word in the settlement vocabulary. They are found through the syntax tree rather than by searching text, because this guard's own comment stripper blanks string contents and would have made that check permanently, silently green. From 0.84.0 the floor is 12.0.1 and the guard witnesses the declarations the package now reads: the per-session background listing method, the `background.listFace` capability bit, and the closed background-status vocabulary that the registry classification switches over exhaustively. Every installed SDK the guard can see at or above the floor must carry those declarations too, so lowering the floor to a line where they do not exist fails. From 0.94.0 the floor is 18.1.0: the shipped declarations name the approval-attribution type and the six-word request permission-mode type that SDK 17 introduced, and the cancellation-cause, status-derived-code and routing-failure values are generated from the same SDK at build time; the guard reads those values on the floor-line SDK it can see. |
 | `scripts/run-engine-caps-ledger-test.mjs` | A per-key disposition ledger for `GET /v1/capabilities`. The SDK's `Capabilities` grew from 74 keys to 93 in one release and nothing on the board could see it: this package consumes that table through four synchronous readers, and *nineteen new positions arriving while the package does not move* is exactly the disease shape this repo keeps logging on other axes — the fact is already on the wire, the package boundary is the cell that swallows it, and no client can read it however they write their side. So the ledger is reconciled **element-wise against the SDK interface in both directions**: a key the SDK added with no ledger row is red (someone must classify it), and a row for a key the SDK removed is red too (a registration that no longer does anything). Each row then has to survive its own claim — a `read` row names the source file, and the **code** there (comments stripped) must really mention the key, because prose asserting an alignment is the classic way these guards go hollow; a `not_read` row must have **zero** read sites in the tree, so wiring one up while the ledger still says the package ignores it is red rather than invisible. The census behind those two directions recognises five call shapes, each of which really occurs here — a reader whose base argument carries its own parentheses, a direct `caps.<key>`, a narrowing cast, an own-property read helper, and a `*_CAP` constant — and proves it on fabricated samples first, since a census that recognises one shape reports "nothing here" for the other four. What the guard deliberately does **not** judge is whether a position *ought* to be read: that is a design call, and the ledger only pins that every capability was looked at once by a person and that what they wrote down does not contradict the code |
 | `scripts/run-sql-engine-capability-test.mjs` | The SQL-posture read face and the four-state capability reader underneath it. One capability cell here carries **four different things**, and each one points an operator somewhere else: nothing has been observed yet in this process (a one-shot doctor run is always in that state), the response arrived but cannot say (no store posture, or one that cannot be read), the engine reports a store posture without an SQL seat — *this deployment has no SQL backend*, which is a **positive fact** rather than an absence — and a full reading. Fold any two together and the screen states something flatly, confidently, and wrongly, so every positive control here is paired with a control pointing the opposite way, and the four sentences the doctor row can print are checked to be pairwise distinct and non-implying. The reading itself is narrowed no tighter than the mint: `txnMode: null` is a **legal value** — two of the three engines always report it that way, and the upstream type note names reading it as "optimistic" as the error — so treating it as malformed would throw away the entire reading for ordinary deployments, which is the same disease this repo logged when a consumer's domain was narrower than the producer's. A response that cannot be parsed **clears** the cell rather than leaving the previous engine's answer in place, and a separate invalidation port exists for the case the generation latch cannot catch — a same-port respawn whose new probe never succeeded, where the stale reading would otherwise be answered as current fact. Untrusted values (the isolation string is read back from a database server variable) are sanitised and bounded before display, and the bound is applied **before** escaping so a visible escape never gets cut in half. Finally the export names are themselves a guard: the shell still carries a copy that is meant to go red on the package's same-named export and be swapped out, so renaming anything here would silently disarm that lock. Current engines (≥7.106.0) moved the SQL posture under the always-present store posture (`store.sql`) and dropped the top-level key, and from 0.92.0 the reader reads `store.sql` alone (absent there means the engine reports no SQL posture): the top-level position of servers below the support floor is no longer read, a response without a readable store posture is "not reported" — the same answer the other readers sharing that witness give — and an unreadable SQL seat clears the cell; the engine's own projection is replayed to prove every form it mints reads back |
-| `scripts/run-web-search-backend-capability-test.mjs` | The deployment-default WebSearch backend read face (`capabilities.webSearch.backend`, engine ≥7.82.1). Same four-state discipline as the SQL and write-protection cells, with two things that are specific here and therefore guarded: a **missing key** (an older engine) and an explicit **`"none"`** (the engine says this deployment has no default search backend) point an operator in opposite directions — "cannot tell" versus "not configured" — and must never be folded; and the `none` sentence has to say both halves of the contract at once: the default scenario mounts no WebSearch tool, **and** a caller-supplied `webSearch` setting can still mount it on a single-user lane, because the capability advertises the deployment default, not whether this request has search. The backend word is read as an **open set** — the engine's closed set is typed from its own provider tuple and grows with it, so hand-copying three words here would turn a newly configured backend into "unreadable" (the narrower-than-the-mint disease this repo already logged once). `webSearch: null` is malformed rather than `none` (the mint never emits `null`), extra members never cross, an unparseable response clears the cell, a stale probe generation is dropped, the invalidation port clears to "not observed", and the open-set word is sanitised and bounded before display |
+| `scripts/run-web-search-backend-capability-test.mjs` | The deployment-default WebSearch backend read face (`capabilities.webSearch.backend`, engine ≥7.82.1). Same four-state discipline as the SQL and write-protection cells, with two things that are specific here and therefore guarded: a **missing key** (an older engine) and an explicit **`"none"`** (the engine says this deployment has no default search backend) point an operator in opposite directions — "cannot tell" versus "not configured" — and must never be folded; and the `none` sentence has to say both halves of the contract at once: the default scenario mounts no WebSearch tool, **and** a caller-supplied `webSearch` setting can still mount it on a single-user lane, because the capability advertises the deployment default, not whether this request has search. The backend word is read as an **open set** — the engine's closed set is typed from its own provider tuple and grows with it, so hand-copying three words here would turn a newly configured backend into "unreadable" (the narrower-than-the-mint disease this repo already logged once). `webSearch: null` is malformed rather than `none` (the mint never emits `null`), extra members never cross, an unparseable response clears the cell, a stale probe's late landing is dropped by the capabilities cache, the invalidation port clears to "not observed", and the open-set word is sanitised and bounded before display |
 | `scripts/run-terminal-cause-projection-test.mjs` | The `7.64.0` wire reshape, projected. A run's ending stopped being eight parallel flat keys and became **one tagged cause** (`completed \| failed \| blocked \| paused`), and a tool call's gate stopped being four orthogonal words and became **one record** (`disposition` / `settlement?` / `origin?`). Both are read in exactly one place in this package, and this guard pins them at **two levels**, because the dangerous seam is "the reader was updated, the consumer was not": each terminal arm is checked on the reader *and* on the `subtype` / `is_error` / `errors[]` the projector actually emits. Two properties carry most of the weight. First, a terminal word this reader does not know is **never** laundered into an empty success — it lands on an `unknown` arm carrying the word verbatim, while a payload with no terminal word at all (the mock lane) keeps the success arm exactly as before, which is the one and only case the reader answers `null`. Second, the three window words (`approval_window_expired`, `denial_limit_window_expired`, `park_sla_expired`) must each be told apart by a different predicate: the previous generation collapsed all three onto one `timeout`, and re-merging them would throw away the discrimination this reshape just restored. Two byte generations are read by one reader, keyed on the discriminator upstream nailed (`"terminal" in result`): the current cause form, and the **flat** form that a current engine still emits on two lanes — replayed persisted bytes, which the service passes through verbatim rather than back-filling, and the service's own rejection envelope. A cause-form payload that also carries stale flat keys must ignore them entirely: keeping one compatibility read is what gives a single fact two sources. The same file also pins the MCP delivery verdict and HTTP status riding the wiring manifest, the four-state write-protection reading (where three of the four states mean *cannot tell*, and none of them may be printed as "there is no table"), and the park-reopen fetch identity: that predicate is asserted through the **real entry point**, since the defect being fixed was precisely a call site wired to a different predicate than the one that routed the row there. From 0.80.0 one of those three boundaries flips: the key naming **who settled a refusal** stopped being a dead byte and became part of the wire, so the check stopped scanning the build output for the word and started reading the request bodies the two decision legs actually send. A refusal attributed to the deployment's own policy carries the word; one attributed to a person, one with no attribution at all, and one carrying a word the vocabulary does not hold carry nothing — the wire has no slot for “a person decided this” other than the key's absence, so inventing one would be minting a word upstream does not have. The allow family never carries it on any of its routes, because that combination is refused before the approval is judged while the side effects of allowing have already landed, and the three refusals nobody was asked about (a card that failed, a user who walked away, an interruption) carry nothing either. A deployment that signs the bodies it accepts does not sign that word, and there is no capability bit to ask beforehand, so a refusal on exactly that ground is answered by re-sending the same decision once with that one key removed — byte-for-byte the same otherwise — rather than letting an optional note take the whole denial down with it. The guard measures that along three axes: the decision still lands and is reported as decided with the attribution handed back and a separate flag saying it never reached the wire; a caller who aborted in between gets no second request; every other refusal code, and every decision that never carried the key, send exactly once. The classification of a second failure is made from what the second body actually carried, not from what the card asked for. From engine `7.104` a synchronous submit that stops at a gate returns the engine result itself plus a three-key receipt: it now carries the tagged cause, so it reads as `paused` on the cause generation (older engines still send the flat three-key body, which keeps reading as before), and the top-level park word is looked at first, the same order the SDK documents for all three generations, so a body the server says is parked is never read as finished. A parked run row now carries its result too, and the headless reconnect path turns it into the parked terminal frame on the first attempt instead of spending its whole retry budget — both generations are pinned, including an end-to-end drive through the public reconnect entry point. When the caller passes the failure's error carrier and it is a session-history tool_use mismatch, the edit-and-resend reading yields to the recovery sentence that line already carries; the carrier is handed to the same mismatch classifier, so the reading and the classifier never disagree on a carrier, whatever its shape; without a carrier the reading is compared byte for byte with the previous release's package. The machine-readable cause of a `blocked` ending is pinned as well: when a deployment hook keeps a prompt from being submitted, the blocked reading carries an optional detail with three arms — a known kind with a known cause word, a known kind with a cause word this version does not know, and a kind this version does not know — and a word outside the tables comes back verbatim and is never folded into a known one. An absent or malformed detail is left out entirely rather than guessed at, only own properties are read (planting the same names on the global prototype changes nothing), a flat-plane row and the other endings have no such position, and a reading without a detail is compared byte for byte with the previous release's package. Real engine and server packages from both generations are driven end to end — the engine's four hook endings, the server's redaction of the hook's own words, and the reader — and a per-generation fact pin says which engine and server versions mint the detail. |
 | `scripts/run-auto-mode-unavailable-test.mjs` | The fact behind "you are being asked because the auto-mode classifier could not run", and the one place its sentence is minted. The cause table is a **copy**, reconciled word for word in both directions against the installed engine's own bytes — it narrowed upstream, and the guard follows rather than keeping the old shape: a table checked against something nobody ships any more is the oldest way for a guard to be green and wrong. The retirement is held from both sides — the removed table must really be gone upstream, and the removed reader and word must really be gone here — while the word that left keeps arriving cleanly from an older engine, because the reader takes the cause as an **open set**: the vocabulary belongs upstream, so a copied list here would discard a legal value the day one is added, and the value discarded is precisely "this outage is a NEW kind". The reader's one exclusion is the word the engine says it never stamps here — the classifier did run and did answer, just outside its contract, so reading it as a failure would invent an event the engine denies. That exclusion used to be derived from a second table which no longer exists; the reason for it never lived in that table, so it is now stated where it actually comes from, pinned as a **named** set (a magic literal scattered through the reader reds) and cross-checked against the engine's own verdict declaration and against the reader having exactly one such comparison. One reader serves both the live ask and its durable parked twin, since the two carry the same key path and a second copy is how two ledgers drift apart. Absence is pinned as absence — most asks never consulted a classifier at all — and the sentences are checked mutually distinct, prototype-safe, and walked end to end: an unknown word reaches the sentence a person reads (the fallback that names it verbatim) and the status reading (unavailable for this round, never a fallback to "available"), with counter-controls proving neither assertion is vacuous |
 | `scripts/run-engine-notice-catalog-test.mjs` | The engine-notice catalog and its audience table. Whether a notice deserves a person's attention is not decided by whether this end happens to have a phrasing for it — that drifts with each client's build order — but by whether the engine minted the code into its own written catalog; the audience row answers the separate question of *who* the fact is for, since an operations fact pushed at an end user is noise and a user-facing fact buried in an operator log is something withheld from the person who could act on it. Both tables are reconciled against the installed engine's own artefacts in both directions and pinned in lockstep with each other, unknown codes fall back to the conservative operator side, and catalog membership is tested on the raw value so a code carrying control characters cannot impersonate a registered one after sanitizing. The reader for a dropped MCP injection keys on its own code alone and treats a missing session, server or reason as absence rather than throwing at a read site. A reverse pin enforces the upstream's single-mint contract: the engine composes those sentences from the host's facts, so a copy of them appearing in this package's source or build is a second source that would drift, and fails. From 0.84.0 it also covers the reader for the two read-directory grant notices: it recognises only those two codes, needs the tool call id to match a card, passes the rejection reason through as written, and treats only the granted notice as evidence that a directory was added; a granted notice without both the directory and the spelling the engine now holds, or with a scope other than `exact`, is not read at all, and the scope word is pinned to the engine's type at compile time. The server also mints a few notices of its own through the same channel; those codes live in a second table with their own audiences, kept apart from the engine mirror (which must stay equal to the engine's catalog) and reconciled against the server's published package when one is supplied, so a user-facing server notice is no longer filed under operations. One dispatcher returns the typed facts for every code that has a reader, discriminated by code and tagged with its audience — only the user-audience codes belong on a user surface — and the guard ties the dispatch table to the module's own exported readers in both directions, so a reader cannot be exported without a row and a row cannot be dropped without the guard failing. From 0.85.0 it also covers a third server-minted notice, the one saying that part of a session's saved history could not be read when the session was reopened: it is a user-audience notice, its two counts are read one by one and anything that is not a non-negative integer reads as `unknown` rather than zero (neither "nothing was skipped" nor "nothing is left" may be invented), and one extra sentence — the context is empty but the turn runs — is given only when the count of entries left is exactly zero. When the server package is supplied, the guard drives the server's own emitter for that notice and reads what it emits back through the dispatcher. From 0.86.0 the catalog grows by six engine notices — a personal rule store that could not be read and was skipped, an auto-mode classification that could not decide and where the call went, a requested permission mode that did not take effect, a permission mode answering a question on the person's behalf, a mode release past a read-deny table, and a legacy checkpoint's mode keys being migrated — each with its own typed reader: branching words (the classifier's cause and destination, the reason a mode did not take effect) are read as closed sets, descriptive words pass through as written, and a malformed notice reads as absent; two of them are checked against notices produced by the engine's own emitters. The unresolvable-ask notice's optional remedy sentence is carried verbatim when it is a non-empty string — never trimmed, rewritten or filtered by cause — and its absence (older engines, other causes, an empty or non-string value) leaves the rest of the view unchanged; a notice without it reads byte-for-byte as before, and the engine's own mint function drives three cases through the dispatcher. A fourth server-minted code, the user-facing notice that a model endpoint could not be connected to, is registered with its audience and has its own reader: the failure code, the endpoint, the remedy sentence and the session are required, the name of the proxy variable is optional, the code and remedy pass through as written, and the server's sentences are never copied into the package. With server fixtures, the server's own decorator mints the notice for a direct and a proxied connection and for every connect-failure code it knows, and stays silent when a response arrived or the call did not end in error; the audience table is reconciled against both fixture generations, a row newer than the older fixture being required to be absent there. From 0.88.0 the catalog grows by one operator notice — a provider request whose tool turn broke the pairing rules was repaired on its way out — with a typed reader: each repair's form word passes through as written (the engine does not publish that word list from its entry point and has announced more words), the call ids are read as a non-empty list, any malformed entry makes the whole notice read as absent, and notices produced by the engine's own repair law and emitter read back field for field. The model-unreachable reader reads only the notice detail's own properties, like the other server-minted readers. From 0.89.0 the engine's own catalog registers those server-minted codes as well — each code carries an audience and a record of who mints it — so the package's separate table for them is retired and both the membership check and the audience lookup answer from the engine's tables alone, the four codes keeping the audiences they had; the guard reconciles the engine's record of server-minted codes against the published server's stream allow-list on the newer server generation and against the older server's own table on the older one, and the generator refuses an engine whose minter table is not in lockstep with its catalog. Two user-facing notices gain typed readers: an incomplete tool turn (cause, count, session and run are required, the call ids are optional and copied, and a cause word the package does not know passes through as written) and user input a run may not have used — one reader for the steer and follow-up codes, the kind decided by the code, the count required, the cause passed through as written and never filled in when absent — plus one user-facing sentence that names only the count, does not claim the input went unused when the server could not tell, and falls back to a cautious sentence for any cause it does not recognise. Notices produced by the engine's and the server's own emitters read back field for field, and every reader answers the same for frozen and unfrozen input. Every reader and the by-code entry point read a notice's code and detail as own properties only: a notice whose code or detail sits on its prototype, or on a polluted global prototype, reads as nothing through all three doors alike. From 0.90.0 the catalog is eighty-five notices: the engine retired the notice for the one-off migration of old parked rows, so the catalog, audience and minter tables lose a row, its fact reader and named type are gone from the public surface, and the code is treated like any code outside the catalog (the dispatcher answers nothing, the catalog check answers false, the audience stays the conservative default). The unused-input reader also reads the child-session field carried by the copy a parent session receives when a sub-agent ends with inputs unused — taken from the engine's real minting site — and the user-facing sentence says so for the two definite shapes, with no identifier in it; every malformed form of the field drops only that field. The register now holds the allow-list notice under its 0.92.0 engine name with the user audience; the earlier engine name is not registered and falls to the operator default, which is the audience it had on the engines that minted it. Since 0.94.0 the user-input-not-used reader takes its lane and count key from the generated lane table rather than from two hand-written codes, and the catalogue expectations follow the installed engine generation. From 0.94.0 the allow-list notice has a fact reader: each entry that reads as nothing comes back with its list source, its machine value exactly as the server sent it (control characters included), a display form produced by the package's single untrusted-text exit, its reason and the optional skill, list, verdict, current-name and value-type positions; a malformed entry list makes the whole reading absent. The model-unreachable reader carries the leg's run id as an optional `taskId` when the server sends one, and drops only that key when it is unusable. |
 | `scripts/run-tool-roster-projection-test.mjs` | The leg's tool roster — what the engine says it actually mounted and what face each tool wears — replacing three word lists that were only ever an estimate taken from one traffic capture against one pinned engine. The reader copies the engine's own all-or-nothing discipline: a roster whose row cannot be read, or whose declared count disagrees with the rows, is dropped whole rather than handed over short, because a consumer reading a short roster concludes the missing tools are not mounted — the upstream says in as many words that this is worse than sending nothing. A malformed *face* on a row (path target, render hints) drops only that face, since a face is not an identity. Shims are built strictly from roster rows and never guessed from a tool's name, and an axis that cannot be read stays absent rather than defaulting to `false` or `never`, which would render "unknown" as "safe". For run-time changes the guard pins the one hard rule in the contract: a digest that does not match is **not** a rejection — the carried roster is the new state regardless and only the summary becomes unusable, because refusing the swap would leave the consumer holding a stale roster forever. One reading here answers a question that the terminal state structurally cannot: whether this run was assembled with any file-and-shell tools at all. The engine's terminal vocabulary says a run finished, not whether the work got done, so an orchestrator that waits for the end and then guesses has nothing to guess from — while the assembly manifest already said it at the start, one row per mounted instance with the single condition that mounted it. The reading is three-state and both folds are refused: a roster that is readable and carries no such row is the engine stating a fact, while no roster at all is not that fact — the static half of a manifest never carries one, and an older engine reports rosters without naming the mount condition at all, where an empty count would be a statement about the reader rather than about the run. Those two are kept apart in the reason the reading carries, and the wording for every unknown case is checked never to claim the run had no tools. The same roster now decides the tool list on the first line of a non-interactive run: the host holds that line until the roster arrives and lists exactly what the engine mounted at the start of the run, in mount order. The guard runs a real assembly frame through the projection into the decision, and pins that the host falls back to the estimate only once the roster is known not to be coming — a manifest without one, an unreadable one, model output or the run's end arriving first — rather than on a timer alone (model activity counts, including a model call that is still waiting or retrying; an error line the stream synthesizes when a run fails before assembly counts as the run ending), that a sub-run's manifest is never mistaken for the run's own, that an empty roster is taken as the engine's answer rather than as silence, and that the wait bound covers both sequential default budgets the engine gives an external tool server to connect and list its tools. The holding logic itself lives in the package as a small per-run gate — buffer, decide once, release the held messages in arrival order, then pass through — and the guard drives real stream output through it to pin that the release happens exactly once, at the manifest, releasing exactly the held prefix. The ordering itself also lives in the package as a stream wrapper, and the guard checks the final output a consumer reads: the first line is always the tool-list line, a message that arrives while that line is still being built comes after it, a timer firing races nothing out of order, a source that ends or fails before the decision still gets its first line and held messages out before the error, and an early exit closes the source. From 0.84.0 the roster-derived sentence source no longer throws on a value it does not recognise, including a reading of the manifest's `hands` section passed by mistake: it answers the same "not stated" sentence as the `hands` reader, from one shared source, and its six known sentences do not change. From 0.86.0 that first line also says where its tool list came from, in two added keys the wrapper writes onto the host's own line object in place: a source of `engine-roster`, `estimate` or `host-static`, and, only for an estimate, which of the four reasons made the roster known not to be coming. The guard checks every decision path end to end, that the host's object keeps its identity and every other key byte for byte, that stale values of the same two keys left by the host are replaced (a stale reason is removed rather than left behind), and that a host line that cannot be written is passed through untouched instead of failing the first line. |
 | `scripts/run-permission-rule-issue-codes-test.mjs` | The rule-lint refusal codes an engine reports when it will not compile a permission rule. The SDK publishes neither a schema nor a type for them, so the package mints the table from the engine's own bytes and the guard pays the cost of that copy instead of leaving it to somebody remembering: it parses the codes the engine actually mints and reconciles them against the table in both directions, so a code added upstream (the user would see a bare code) and a code only the package believes in (a branch that can never fire) both fail. It also reconciles the table plus a small retired ledger against the engine's declared union, which is deliberately not the same set — one member was renamed and its old name is still declared — so reviving a code the engine will never mint again is impossible and a future stale member shows up immediately. Sentences are pinned one per code, mutually distinct, and split by family: a rule that is wrong and a rule that is legal but unsupported on this lane are different next steps and may not share a sentence. The engine's own message rides along as prose — sanitized and capped after escaping, never matched on The set of codes the engine actually mints now also counts declared codes minted outside the rule compiler (the allow layer reports an entry that is not a string with one). Codes an engine release retired but older engines still mint are found by comparing the receiving server's bundled engine with the installed one rather than by a hand-kept list; each keeps the exact sentence it had before, and the codes that replaced them each get their own sentence with no machine identifiers in it. Since 0.94.0 the base-path refusal sentence is checked to hold on both engine generations (base missing, or declared but not a usable absolute path), with the premise run on the installed engine. |
-| `scripts/run-gate-vocabulary-test.mjs` | The two gate vocabularies — who denied a call (`DeniedBy`, the installed engine's ten words) and who asked about it (`AskOrigin`, eleven) — together with the one place their sentences are minted, so the same denial does not read three different ways across three clients. The tables are copies, not opinions: the asker table is reconciled in both directions against the lists the installed SDK and engine packages publish, and the denier table is exactly the installed engine's own list (from 0.92.0 the one word only engines below the support floor still sent is no longer in it), checked so that every word the SDK declares is present except that retired word, and every word ahead of the SDK comes from the engine — a word added upstream (nobody renders it, the user sees a bare code) and a word only the package believes in (a branch that can never fire) both fail. Every word must carry its own literal sentence and no two may collide, including the sibling pairs the upstream deliberately split apart — an organization store and a personal rule store being unreadable send you to different people, and the two tighten origins exist precisely to name which layer of engine logic asked. The two fallbacks are pinned distinct because an unknown word means different things in each: for a denial layer this build does not know, the sentence says only that this build cannot say which layer refused the call or why — it no longer guesses at a newer engine or a damaged record, since the most likely unknown word is the retired one replayed from an old session; the asker vocabulary is genuinely open (the server only checks for a non-empty string, so an unknown word just means the client is older than the engine). Alongside them sits an **uplift anchor** rather than a third table: the reason a call was decided the way it was is a distinct semantic face from who denied it and who asked, one upstream has not mirrored into the SDK at all, and one whose newest member — a shell command allowed because it only reads — has no sentence anywhere yet. Minting the union here would create the second drifting source the day upstream publishes it, so the guard instead asserts the **absence** from both ends: the SDK declarations carry no such union near that word, and the installed engine’s own list does not carry the word either. The engine end fires first, on the batch that raises the dependency, which is exactly when the ownership question should be answered; the SDK end fires when the mirror lands. Either red is the work order to mint the sentence, never a reason to delete the anchor. A fourth mint now sits beside the three tables and is not a table at all: a single presence-only fact — that no saved rule and no standing posture can retire this question — earns one sentence, taking no argument precisely so a caller cannot mistake it for a second kind of mandate, pinned distinct from every sentence the tables mint, pinned never to point at rule-writing, and pinned not to overclaim the stronger neighbouring demand that a person rather than a configuration must answer; it must not say the question is asked every time — an answer for this one call may come from the person, a hook or an automatic check the deployment runs — and its wording is checked against the engine package's own description of the mandate. A fifth table joins them from 0.80.0: the words for **how a wait ended** (thirteen from 0.80.0, fourteen from 0.90.0), mirrored in both directions from the engine's own declarations — the table's owner — with the wire SDK's copy held alongside as a second witness that must match it word for word and in order, so the day the SDK falls a generation behind, that is what turns red rather than the mirror silently following the wrong source. The newest of them says a deployment's own policy answered the card — not a person, and not “nobody could be asked” — so the guard pins it apart from both neighbours by behaviour, feeding every one of the words through all five named predicates and checking which word makes which one speak, rather than what any predicate returns. Two of them also decide how a refusal is filed in the session transcript; that mapping is minted once and reused by both of the package's own entry points, and anything outside those two words yields nothing rather than a guess. Since 0.83.2 a sixth list covers the word a mandated question stands on (`APPROVAL_MANDATE_WORDS`, six words): it must equal the engine's own list word for word and in order, membership is exact, the card reader `readApprovalMandate` answers only for an own key holding one of the six words, and each word has one fixed sentence explaining why the question must be confirmed — six distinct sentences that never point the reader at writing a rule, never promise a question every time, never claim only a person may answer, and repeat no other sentence the package mints. The list is also pinned against the engine's type at compile time in both directions, while the published build references no engine package at all: every `.js` and `.d.ts` file in the build is scanned, and the same scan is first shown to fire on references planted in a scratch directory. From 0.90.0 the settlement table carries `policy_allowed` (a machine allow, listed before `policy_refused`) and the mirror is checked against the installed engine byte for byte, in order; the installed SDK (12.x) predates the word, so its list is compared as "the engine's list minus one registered lagging word", and every SDK witness at 16.0.0 or later must carry exactly the engine's set of words. The party that settled is the engine's five-word set, read open (any word, including an unknown one, comes back as written); a policy allow is no person's approval and no refusal (all five named predicates stay silent on it, and it files no denial class in the transcript), and the refusal word reads the same under both party spellings of the two server generations. The one lagging word that loosens the compile-time check against the installed SDK is itself guarded: it must stay a non-empty union of string literals (widening it to `string` compiles cleanly, so only this check catches it), the check's declared type must stay exactly as written, the word list must equal the list this check registers, and every lagging word must be absent from the SDK at the peer floor yet present in the newest witness SDK and in the installed engine's list. |
+| `scripts/run-gate-vocabulary-test.mjs` | The two gate vocabularies — who denied a call (`DeniedBy`, the installed engine's ten words) and who asked about it (`AskOrigin`, eleven) — together with the one place their sentences are minted, so the same denial does not read three different ways across three clients. The tables are copies, not opinions: the asker table is reconciled in both directions against the lists the installed SDK and engine packages publish, and the denier table is exactly the installed engine's own list (from 0.92.0 the one word only engines below the support floor still sent is no longer in it), checked so that every word the SDK declares is present except that retired word, and every word ahead of the SDK comes from the engine — a word added upstream (nobody renders it, the user sees a bare code) and a word only the package believes in (a branch that can never fire) both fail. Every word must carry its own literal sentence and no two may collide, including the sibling pairs the upstream deliberately split apart — an organization store and a personal rule store being unreadable send you to different people, and the two tighten origins exist precisely to name which layer of engine logic asked. The two fallbacks are pinned distinct because an unknown word means different things in each: for a denial layer this build does not know, the sentence says only that this build cannot say which layer refused the call or why — it no longer guesses at a newer engine or a damaged record, since the most likely unknown word is the retired one replayed from an old session; the asker vocabulary is genuinely open (the server only checks for a non-empty string, so an unknown word just means the client is older than the engine). Alongside them sits an **uplift anchor** rather than a third table: the reason a call was decided the way it was is a distinct semantic face from who denied it and who asked, one upstream has not mirrored into the SDK at all, and one whose newest member — a shell command allowed because it only reads — has no sentence anywhere yet. Minting the union here would create the second drifting source the day upstream publishes it, so the guard instead asserts the **absence** from both ends: the SDK declarations carry no such union near that word, and the installed engine’s own list does not carry the word either. The engine end fires first, on the batch that raises the dependency, which is exactly when the ownership question should be answered; the SDK end fires when the mirror lands. Either red is the work order to mint the sentence, never a reason to delete the anchor. A fourth mint now sits beside the three tables and is not a table at all: a single presence-only fact — that no saved rule and no standing posture can retire this question — earns one sentence, taking no argument precisely so a caller cannot mistake it for a second kind of mandate, pinned distinct from every sentence the tables mint, pinned never to point at rule-writing, and pinned not to overclaim the stronger neighbouring demand that a person rather than a configuration must answer; it must not say the question is asked every time — an answer for this one call may come from the person, a hook or an automatic check the deployment runs — and its wording is checked against the engine package's own description of the mandate. A fifth table joins them from 0.80.0: the words for **how a wait ended** (thirteen from 0.80.0, fourteen from 0.90.0), mirrored in both directions from the engine's own declarations — the table's owner — with the wire SDK's copy held alongside as a second witness that must match it word for word and in order, so the day the SDK falls a generation behind, that is what turns red rather than the mirror silently following the wrong source. The newest of them says a deployment's own policy answered the card — not a person, and not “nobody could be asked” — so the guard pins it apart from both neighbours by behaviour, feeding every one of the words through all five named predicates and checking which word makes which one speak, rather than what any predicate returns. Two of them also decide how a refusal is filed in the session transcript; that mapping is minted once and reused by both of the package's own entry points, and anything outside those two words yields nothing rather than a guess. Since 0.83.2 a sixth list covers the word a mandated question stands on (`APPROVAL_MANDATE_WORDS`, six words): it must equal the engine's own list word for word and in order, membership is exact, the card reader `readApprovalMandate` answers only for an own key holding one of the six words, and each word has one fixed sentence explaining why the question must be confirmed — six distinct sentences that never point the reader at writing a rule, never promise a question every time, never claim only a person may answer, and repeat no other sentence the package mints. The list is also pinned against the engine's type at compile time in both directions, while the published build references no engine package at all: every `.js` and `.d.ts` file in the build is scanned, and the same scan is first shown to fire on references planted in a scratch directory. From 0.90.0 the settlement table carries `policy_allowed` (a machine allow, listed before `policy_refused`) and the mirror is checked against the installed engine byte for byte, in order; the installed SDK (12.x) predates the word, so its list is compared as "the engine's list minus one registered lagging word", and every SDK witness at 16.0.0 or later must carry exactly the engine's set of words. The party that settled is the engine's five-word set, read open (any word, including an unknown one, comes back as written); a policy allow is no person's approval and no refusal (all five named predicates stay silent on it, and it files no denial class in the transcript), and the refusal word reads the same under both party spellings of the two server generations. The one lagging word that loosens the compile-time check against the installed SDK is itself guarded: it must stay a non-empty union of string literals (widening it to `string` compiles cleanly, so only this check catches it), the check's declared type must stay exactly as written, the word list must equal the list this check registers, and every lagging word must be absent from the SDK at the peer floor yet present in the newest witness SDK and in the installed engine's list. The origin word on a gate record and on a released-by-mode record is read as an own key only, and a structural pin holds that every source file reading an ask origin goes through the one reader. |
 | `scripts/run-engine-identity-test.mjs` | The engine generation anchors on `/health` (`pid`, `instanceId`, `startedAt`; engine >=7.67.0). `/health` is the one unauthenticated door and its heartbeat is always green, so "another host restarted the shared engine" used to be discoverable only by having some authenticated request hit a 401 first — a path that misreads a restart as a network fault. The reader narrows each anchor independently (one malformed field never hides the other two) and always hands back a reading object rather than an absence, because the caller is asking which anchors answered, not whether there was a response. The comparison is a three-word verdict, not a boolean: `unknown` when the two readings share no comparable anchor at all — an empty intersection means nothing could be compared, never that nothing changed — and the boolean convenience is pinned so that only `true` is an assertion. Any comparable anchor differing decides `changed`, so a reading whose `startedAt` matches while its `instanceId` does not cannot be waved through as the same life; precedence only decides which anchor gets named in the diagnosis |
 | `scripts/run-posture-knob-projection-test.mjs` | The three deployment knobs on the operator face (`serverGates.durableApproval` / `streamAskWindowMs` / `sessionAutoTitle`, engine >=7.67.0), each read as a value **plus who set it plus one operator-facing pointer** rather than a bare value — a bare boolean cannot answer why this particular machine is on this setting or how to pin it back, and a default that flips with the deployment shape is invisible without that. A worker too old to report readings still sends a bare boolean; the reader folds it into the same shell so consumers keep one branch, but raises a `legacy` bit, answers `undefined` from the machine-readable source accessor, and mints a sentence that contains no source word at all — claiming a source nobody reported is worse than admitting the worker cannot say. The other two knobs are honestly absent on such a worker rather than defaulted, a malformed side knob drops only itself while the anchor knob drops the whole reading, and the four sentences are pinned literally distinct so an operator can tell "not observed" from "not reported" from a real value. The last leg reads the installed SDK's `openapi.yaml` and `types.d.ts` directly, including a pin that exactly one knob on this face is numeric — the premise the millisecond-to-prose rendering rests on |
-| `scripts/run-terminal-facts-projection-test.mjs` | The four unconsumed terminal-receipt facts: `TaskResult.effectiveReasoning` / `effectiveMemoryScopes` are narrowed into `_sema_effective_reasoning` / `_sema_effective_memory_scopes` on the CC-shaped `result` (success and error envelopes alike; a malformed value mints nothing, never a default tier), the resume **reopen** family (`resume.env_failed` / `tool_unavailable` / `tool_contract_mismatch`) is a frozen closed set with a reader and three-sentence copy that is disjoint from the refusal and retry-later sets, and `routePairingVerdict` reads `ModelInfo.routePairing` as ok / broken / unknown without policing the open set. A fifth section pins the structured-output key on the success result: the CC-spelled `structured_output` is the only home for the value the wire calls `structuredOutput`. The camelCase spelling this package used to mint on its own — a misspelling of the CC field, not an additive field of our own — rode alongside it for exactly one release (0.79.1) and is **absent from 0.80.0 on**, pinned both by own-key and by `in`, so a consumer still reading the old name sees `undefined` rather than a stale copy. The wire position is read exactly once, so a value-changing accessor is only ever asked for its first answer; absence stays absence; a wire key that is present but `undefined` mints nothing, since a key whose value is `undefined` makes a consumer that tests presence read "the engine produced nothing" as "the engine produced an empty result"; falsy-but-present values such as `null`, `0`, `""` and `false` are still minted, and so are shapes that are not records at all — an empty array, a populated array, a string, a number, a boolean — each carried through by the same reference, because the shape of that value is decided by the caller's own schema and the package does not get to filter it; and the error envelope carries no such key, because the CC error arm has no such field. Which spelling CC itself declares is witnessed from the mirror's own syntax tree rather than a constant copied into the guard, so the day that field is renamed upstream the guard says so. A seventh section covers the two keys the memory-scope reading used to leave behind: the mounted arm's `projectKey` (exactly one of `{ scope }` or `{ absent }`; a missing key means the deployment declared nothing, which is not the same as having looked and found none, and the reason word is read as an open set so a word added upstream is never dropped) and each row's `shadowed` (a positive integer only, never on the write home's own row, never a zero or a coerced value). A malformed or unreadable value mints that one key and nothing else, so a throwing accessor never takes the arm down; the other arms never carry either key; and inputs without the two keys read byte for byte as the previous release does, checked against that release's published artifact on a large generated pool and on a pool of live objects (throwing accessors, revoked proxies) where even throwing versus not throwing must match. The upstream facts are pinned on every generation of server fixture the guard is given: the mint sites and declarations in the engine each one bundles, the server's terminal shaping run for real so both keys cross the wire unchanged, and the server's own project-identity minting fed into the reader. The SDK types are pinned as not modelling the face at all, and the two keys that were declined are pinned as having no mint site on the wire, so the day either changes the guard goes red instead of the decision going stale. |
+| `scripts/run-terminal-facts-projection-test.mjs` | The four unconsumed terminal-receipt facts: `TaskResult.effectiveReasoning` / `effectiveMemoryScopes` are narrowed into `_sema_effective_reasoning` / `_sema_effective_memory_scopes` on the CC-shaped `result` (success and error envelopes alike; a malformed value mints nothing, never a default tier), the resume **reopen** family (`resume.env_failed` / `tool_unavailable` / `tool_contract_mismatch`) is a frozen closed set with a reader and three-sentence copy that is disjoint from the refusal and retry-later sets, and `routePairingVerdict` reads `ModelInfo.routePairing` as ok / broken / unknown without policing the open set. A fifth section pins the structured-output key on the success result: the CC-spelled `structured_output` is the only home for the value the wire calls `structuredOutput`. The camelCase spelling this package used to mint on its own — a misspelling of the CC field, not an additive field of our own — rode alongside it for exactly one release (0.79.1) and is **absent from 0.80.0 on**, pinned both by own-key and by `in`, so a consumer still reading the old name sees `undefined` rather than a stale copy. The wire position is read exactly once, so a value-changing accessor is only ever asked for its first answer; absence stays absence; a wire key that is present but `undefined` mints nothing, since a key whose value is `undefined` makes a consumer that tests presence read "the engine produced nothing" as "the engine produced an empty result"; falsy-but-present values such as `null`, `0`, `""` and `false` are still minted, and so are shapes that are not records at all — an empty array, a populated array, a string, a number, a boolean — each carried through by the same reference, because the shape of that value is decided by the caller's own schema and the package does not get to filter it; and the error envelope carries no such key, because the CC error arm has no such field. Which spelling CC itself declares is witnessed from the mirror's own syntax tree rather than a constant copied into the guard, so the day that field is renamed upstream the guard says so. A seventh section covers the two keys the memory-scope reading used to leave behind: the mounted arm's `projectKey` (exactly one of `{ scope }` or `{ absent }`; a missing key means the deployment declared nothing, which is not the same as having looked and found none, and the reason word is read as an open set so a word added upstream is never dropped) and each row's `shadowed` (a positive integer only, never on the write home's own row, never a zero or a coerced value). A malformed or unreadable value mints that one key and nothing else, so a throwing accessor never takes the arm down; the other arms never carry either key; and inputs without the two keys read byte for byte as the previous release does, checked against that release's published artifact on a large generated pool and on a pool of live objects (throwing accessors, revoked proxies) where even throwing versus not throwing must match. The upstream facts are pinned on every generation of server fixture the guard is given: the mint sites and declarations in the engine each one bundles, the server's terminal shaping run for real so both keys cross the wire unchanged, and the server's own project-identity minting fed into the reader. The SDK types are pinned as not modelling the face at all, and the two keys that were declined are pinned as having no mint site on the wire, so the day either changes the guard goes red instead of the decision going stale. The key set compared across the success and error envelopes is read from the type declaration of the shared fact parts by the compiler's checker rather than from a hand-kept list or a name prefix: every declared key must have an input fixture and every fixture a declared key, and a planted extra key or a removed key is shown to change the derived set. |
 | `scripts/run-export-liveness-test.mjs` | Every runtime export in the public baseline must be **alive**: referenced by some gate, or explicitly registered in `scripts/export-liveness.json` as `contract` (consumed by a client with no gate yet), `internal` (an internal helper amplified onto the public surface by `export *`), `candidate` (with ticket + retire-by) or `retire` (dead; retire-by version). Registration is accounting, not exemption: a row for a name a gate already references is stale and must go, a row for a name no longer exported is red, `retire`/`candidate` rows go red the moment `package.json` reaches their retire-by version, and the row count only ratchets down. When the client repositories are on disk the consumption evidence is checked by name, from committed history and never from a working tree: each client's `origin/main` (or its HEAD when there is no such ref), the checked-out HEAD however old it is, and every local branch head committed within the last 14 days, so an import that so far exists only on an unmerged branch still blocks the retirement. Two readings are kept apart. The registry states facts, so a `contract` row's claimed consumers must equal the clients that really bind the name — a named import from the package or one of its subpaths, a re-export from it, a member read on a namespace or dynamically imported module object, a destructured dynamic import — and an `internal` name that a client binds must be re-registered as `contract`; a client-side declaration that merely shares the name is not consumption. The retirement check asks only whether removing the name could break a client, so it also counts a plain identifier match in any file importing the package, and, when a client re-exports the whole package, any appearance of the name in that client's sources; the guard prints both counts. The guard prints the ref, commit and commit date of everything it read, and it shares one repository reader with the same-name shadow guard, so both guards judge the same snapshot of a client. A client that is absent or is not a git repository is reported as a skipped section, never as a client with no consumers, and a dangling or unreadable branch ref is a fault rather than a branch quietly left out. The reader proves itself on a throwaway repository: an import on a recent branch that is not checked out must be found and attributed to its branch and line, while one on a branch older than the window, in an uncommitted file, or outside the scanned folders must not count. Names that have already left the surface are kept in a per-version `removed` ledger: they must never reappear in the baseline or the registry, and the ledger's versions must not run ahead of the changelog. A name may come back only through a separate `revived` ledger that names it, the version it left in, the version it returns in (later than the one it left in and not ahead of the changelog), a ticket and a reason, and only if it is really back in the baseline; the `removed` ledger itself is never rewritten, and the documentation freshness check reads the same `revived` ledger so a returned name is no longer treated as retired. The `removed` ledger is also checked against history: compared with the same file at the tag of the expected previous release (the newest released version in the documentation freshness check's frozen ledger), every version it listed must still be there, with every name it listed still under that same version — entries can only be added. Only that tag is used: outside a git checkout, when that tag is not present locally, or when it does not carry the ledger, the check is reported as skipped rather than passed, never run against some other tag. Names cannot leave the public surface unrecorded either: every name in the baseline at that same previous-release tag that is no longer in today's baseline must be listed in the `removed` ledger, under the version of the topmost changelog section or under an earlier version as long as it is not a name that has since returned; the check compares the two sets of names rather than their counts, and it is reported as skipped, never passed, under the same conditions. |
 | `scripts/run-wire-refusal-copy-test.mjs` | Two wire refusals read the same way on every client: a cancel's 409 carries one of two codes with opposite dispositions (`conflict.approval_settled` — someone else already decided, go read the result; `conflict.run_not_running` — nothing changed, send the cancel again), an unrecognised or codeless 409 is reported as such rather than guessed, and the submit-side 429 `usage.window_exhausted` is read as a waitable refusal whose wait is stated only when the engine supplied one. `ControlRouter.cancel` raises a distinct safety code for the retry-directly case. A third family covers the refusals that a deny's *settler note* can draw: a deployment that signs the decisions it accepts but does not sign that note, a body whose decision and note contradict each other, and a word the engine does not recognise. All three are refused **before** the approval is judged, so each sentence states plainly that nothing was decided and the approval is still waiting, and each names a different next step — none of them “send it again unchanged”, which would simply be refused again. Two of the three codes already mean something else in this package: one is shared with the plan-review leg, so the reader refuses to claim it unless the caller states that the note really was sent, and the other is split by the field the engine names, because without that field the same code means a review outcome was rejected for its content. The third sentence deliberately does not say the word was misspelled: upstream mints that same field for at least four different reasons, so the only thing it proves is that the engine would not take the settlement details and named which part — which is what the sentence says, and where it points. The three sentences are pinned verbatim rather than by keyword, because a keyword check passes a sentence that tells the reader to send the same body again unchanged, which is the one next step that is certainly wrong. The field itself is read from the bag the SDK keeps additional response keys in, not off the top of the error, since only the hand-built shapes a test would write carry it there. Both decision legs hand the reading back on their outcome, and a leg that never sent the note claims nothing. When a decision fails because the engine cannot read a stored row, the row the engine names is read (from the error's top level, or from the SDK's bag of additional response keys where the real SDK puts it) and carried into the failure reason on all three decision legs, escaped for display.. A billed request refused because the deployment enforces budgets fail-closed and could not confirm the caller's budget with its quota service (`fleet.lease_unavailable`, 503, engine ≥7.106.0) is recognised by its code alone and gets its own sentence instead of the generic server-failure line: the budget could not be confirmed, the request was not accepted and is not at fault, each known cause adds its own clause, and the wait is stated only when the engine supplied one; the turn-error classifier marks the same case, and the engine's own refusal mapping is replayed through the real SDK to prove every cause reads back The router's `cancel_lost_race` refusal carries the same sentence as the reader for its kind (an approval already settled elsewhere, or a conflict the engine did not explain), and there is only one judge of whether a cancel failure is that 409. |
 | `scripts/run-tool-disclosure-progress-projection-test.mjs` | The two wire arms sdk 9.6.0 adds — `tool_disclosure` (name-only tool census: open-set `policy`, `thresholdPercent` absent ≠ default, `deferred`/`activated` full snapshots) and `tool_progress` (one frame, two beats: Bash ticks carry an output tail with `totalLines`/`totalBytes` that come and go together; other tools carry only `elapsedSeconds`) — project to neutral internal arms plus chrome arms. Required keys missing ⇒ `malformed`; bad optional keys drop only themselves; the sub-flow three-key gate keeps child frames off the leader lane; both arms are `required: false` in the arm table with duties stated (the output tail is untrusted raw and must never be fed back to the model). |
 | `scripts/run-mcp-panel-projection-test.mjs` | The `GET /v1/sessions/:id/mcp` panel reader (`projectMcpPanel`; server >=7.77.0 adds the optional `lastLegMcp` key) and the single wording mint for its "last leg" line. Absence of `lastLegMcp` is one literal sentence that never blames the engine version (a new session, a leg outside the retention window, a leg without a manifest and an older engine all look the same on the wire); a key that is present but unreadable is a different sentence plus a `lastLegMcpUnreadable: true` mark, never folded into absence. The `mcp[]` roster goes through the same reader as the live `wiring_manifest` third section, so a replayed roster and a live one have one shape. The two faces of the panel (`servers[]` and the last-leg roster) may legitimately differ, so the view carries no agreement flag and none of the five sentences mentions `servers`. Required keys are pinned to the SDK `openapi.yaml` component bytes **0.69.0:** `fetchMcpPanel` fetches the panel through the SDK client's own `sessions.mcp` call (same transport and auth as every other read) and projects it; transport failure, an unreadable body and an empty session id all come back as `undefined`, never as a fabricated empty panel 0.71.0 adds section K: `mcpEngineLegPresence(view)` — the engine-side MCP presence tri-state read only off the panel view (`unknown` when the view could not be read, never rendered as "no MCP configured") |
 | `scripts/run-absence-fold-census-test.mjs` | A package-wide census of the "absence folded into a positive outcome" defect shape, so that fixing the six sites this release does not merely move the shape somewhere else. The defect is defined by position, not syntax: a fallback position (the unconditional tail return, the `default:` arm, the literal minted when there is nothing to pass on, the value returned from an error path) may only say `unknown` or stay absent, never a positive word. Detection walks the syntax tree of every source file, so comments, strings and multi-line spellings cannot hide or fake a hit, and covers five forms: the right arm of `??` / `\|\|`, the else arm of a ternary, the first return of an explicit `default:`, a `catch` block or `.catch(() => …)` arrow returning a healthy value, and a function whose last statement returns a positive word after other returns. Every remaining hit must be registered with a written reason, an unregistered hit fails the gate naming the file and line, the registered count must equal the real count so a cleared site cannot leave a spare allowance behind, and the gate proves its own teeth behind a fence (a failed self-proof refuses to report any count): each form injected into an in-memory copy must add exactly one hit, two correct spellings are pinned as non-hits, and samples inside comments or strings do not count. It also pins the headline site: the fleet panel projection no longer mints an `end` with `isError: false` on absence |
-| `scripts/run-device-executor-management-capability-test.mjs` | The engine's device-lane self-description (`capabilities.deviceExecutor`, with its management bit from engine ≥7.88.0), read by one reader the same four-state way as its sibling capability readers: a lane object is reported as present and carries the management bit as one of three literal states (on, off, or absent — an older engine really ships the lane object without the key, and that is never folded into `false`) plus three optional lane facts (protocol version as a finite number, in-flight cap per device as a finite non-negative number, socket path as a non-empty string), each read on its own so a missing or malformed one only drops that fact; the lane object must carry `enabled: true` as the contract says, and anything else is an unreadable shape rather than an invented disabled-lane state; an absent `deviceExecutor` key is not reported, `deviceExecutor: false` is the lane being absent, presence is judged by own-property not truthiness, every key is read once, the tee never throws and drops stale generations, and the package owns two verdicts — whether the lane is open (`yes` for any lane object, `no` for the lane being absent, otherwise `unknown`) and whether the `/v1/devices` management verbs are usable (`yes` only when the bit is on, `no` when it is off or the lane is absent, otherwise `unknown`) — while the installed SDK's declaration of the lane object is pinned |
+| `scripts/run-device-executor-management-capability-test.mjs` | The engine's device-lane self-description (`capabilities.deviceExecutor`, with its management bit from engine ≥7.88.0), read by one reader the same four-state way as its sibling capability readers: a lane object is reported as present and carries the management bit as one of three literal states (on, off, or absent — an older engine really ships the lane object without the key, and that is never folded into `false`) plus three optional lane facts (protocol version as a finite number, in-flight cap per device as a finite non-negative number, socket path as a non-empty string), each read on its own so a missing or malformed one only drops that fact; the lane object must carry `enabled: true` as the contract says, and anything else is an unreadable shape rather than an invented disabled-lane state; an absent `deviceExecutor` key is not reported, `deviceExecutor: false` is the lane being absent, presence is judged by own-property not truthiness, every key is read as an own data property (accessors are never run), the retired tee port is a no-op that never throws, and the package owns two verdicts — whether the lane is open (`yes` for any lane object, `no` for the lane being absent, otherwise `unknown`) and whether the `/v1/devices` management verbs are usable (`yes` only when the bit is on, `no` when it is off or the lane is absent, otherwise `unknown`) — while the installed SDK's declaration of the lane object is pinned |
 | `scripts/run-run-cancel-context-test.mjs` | The run record's `cancelContext` side-note (engine ≥7.87.3) read structurally, and the cause of a `turn_aborted{engine_error}` classified from machine-readable evidence only: `cancelled` (code `cancelled`, with the cancel-time context when present) / `engine_error` (any other failure code, passed through verbatim) / `run_still_live` (the record is not terminal — a dropped stream is a client-side fact, not the run's cause) / `unknown` (never guessed). An absent `cancelContext` reads as *not reported*, never as "not cancelled"; `elapsedMs` is never folded to 0. |
 | `scripts/run-suspended-reopen-projection-test.mjs` | The durable `suspended` event's `reopened` key read as three distinct states — `reopened` (with the engine's code, verbatim), `not_reopened` (an explicit `null`), `unstated` (key absent or unreadable) — and carried on the HITL bridge's active gate (`currentGateReopen()`), re-read on every `suspended` and cleared with the gate. The stream driver also turns a `suspended` frame that reads `reopened` into the chrome arm `suspended_reopened` (the reading, the single-source reopen sentence when the code is in the reopen family, the frame's event id, and the run id only when the frame itself carries one), pinned end to end: one notice per event id on a stream, emitted in stream order on a replay from the first frame, nothing for a cursor that starts after it, and an unchanged transcript when the host has no chrome sink or the sink fails. |
 | `scripts/run-panel-identity-normalization-test.mjs` | One background subagent has two ids on the wire — the fleet row id tail and the `task_progress` task id (its transcript id). Every panel event goes through one funnel that rewrites the `tick` / `end` task id onto the fleet row's id once a `fleet-row` has registered the key (`transcriptId` first, `parentToolCallId` as the fallback), carrying the original as `wireTaskId` and marking `taskIdOrigin`; an unbound tick whose row has not arrived yet waits one beat (bounded) and is released verbatim on the next tick / `end`, when the buffer is full, or after `MAX_HELD_WIRE_TICK_BEATS` other fleet-row / end / sweep events (a `sweep` itself leaves it alone: there is no row to settle yet); a normalized `end` that carries no cycle identity borrows the registering row's, so a late close of a revived task is recognized as stale; the key table is an LRU (a task that keeps ticking is never evicted by newer registrations); the residency mark migrates with the id and both keys are cleared on settle — except that a stale (previous-cycle) terminal never clears the revived row's mark — so the notification lane can clear it. Once a tick has been delivered verbatim under its UUID, that UUID is the subagent's key: later fleet rows and fleet-side ends are rewritten onto it (the tail kept in `wireTaskId`), so a consumer sees one row in every arrival order; a late tick from a previous cycle is dropped rather than folded into the revived row. |
@@ -329,33 +329,33 @@ guard still cross-checks the table by name).
 | `scripts/run-memory-spec-wire-test.mjs` | The per-agent **memory spec** (`agents[].memory`) read once for every client, plus the judge for the engine's **closed** key list. Two states are kept apart that clients habitually collapse: an absent `scopes` means *no layers were specified*, never "zero layers", and an explicit `writeScope: null` is a positive fact — this run has memory **read-only** (no remember tool, no consolidation write; recall still works) — which is neither "unspecified" nor "memory off". Each of the four keys is read once, on own properties only (an inherited key never reaches the wire, so reading one would report a value the engine cannot see), and a key that is present but unreadable stays in its own slot instead of collapsing into "unspecified"; `enabled` must be a strict boolean and `scopeContract` is an open-set verbatim word. A spec that cannot be read at all answers *undefined*, kept distinct from an agent that simply has no spec. The judge earns its keep on the consequence: the engine checks this spec against a closed list, so one unlisted key — most often the retired singular `scope` — is refused together with the **whole agent definition**, not just that key, and the single sentence minted here says so. What counts as "on the wire" is decided by the bytes, not by the shape of the in-process object: both the reader and the judge work off a `JSON` snapshot of the spec taken **in its property position** (wrapped under the same key, never serialized as a root value — otherwise a `toJSON(key)` that branches on the key hands us one shape and the engine another: one such input made the snapshot say *read-only, no violations* while the real bytes carried the retired key and a writable scope), because `Object.keys` and `hasOwn` disagree with the serializer in ways that change the answer — a non-enumerable `writeScope: null` would otherwise be reported as "memory is read-only for this run" while the engine receives *unspecified* and may still write; a key whose value is `undefined` would be reported as a violation that never leaves the process; a `toJSON` (even inherited) adds keys that `Object.keys` cannot see, including the retired singular one; and a throwing getter would let the judge claim it had looked when the spec cannot be serialized at all. A spec that fails to serialize is reported as unreadable by both ports, and the snapshot is taken once, so every getter runs exactly once. The judge answers in three states, never two: `[]` is an assertion (*looked, nothing unlisted* — including an agent that carries no spec at all), a non-empty list is what it saw, and *undefined* means it could not read the spec (a non-object item, an array, an unreadable `memory`, a throwing getter) — an unreadable spec never poses as a clean one, and an array is not a spec so its index keys are noise rather than findings. It reports only the snapshot's string keys, sorted and bounded, so a prototype, symbol, non-enumerable or `undefined`-valued key is never blamed while a `__proto__` that really does serialize is; the empty-string key is kept rather than dismissed as noise, because it does serialize and dropping it left a non-empty violation list with nothing said about the consequence; every key name in the sentence is quoted and escaped one code point at a time, so no escape is ever cut in half (a half-cut escape used to make the closing quote itself look escaped) and an empty name, a key literally named `""`, a key containing a backslash and a real control character versus a literal `\uXXXX` all read as different violations; a name too long to show is marked `(truncated)` outside the quotes with a pointer to the judge's verbatim list, so a prefix is never presented as the whole key — two long names sharing a prefix do show the same, which is why the mark and the pointer are there; key names are sanitized and bounded on the way into the sentence while the judge itself hands back the verbatim key, because sanitizing belongs in prose and never in a verdict. The announced future key `projectKey` is still unlisted today and is reported as such, with a sentence saying it is not a typo. The two construction-time refusals (`config.memory_project_key_spelling` — a spelling, 400; `config.memory_write_scope_mismatch` — a conflict with the scope already in force, 409) join the existing `config.` recognition table rather than a second word list, and each gets one sentence stating that the refusal landed **before the run started**, so nothing ran; the engine owns the triage and an unrecognised code gets no sentence at all. The write face is widened in the same batch so the package can actually mint what the reader can read: `TaskAgentWireMemory.writeScope` is now an optional `string \| null`, since a reader that understands "memory is read-only for this run" while the writer cannot express it is worse than no reader at all — it makes the support look real. Minting `null` survives serialization and reads back as read-only, minting `undefined` drops the key and reads back as unspecified, and the projector still pins `writeScope` explicitly every time. The accepted key list is reconciled against an upstream witness rather than a second local copy: the guard reads the SDK's own declaration comment for this key, requires the two sets to match in both directions, requires that comment to still name the singular `scope` as retired, and requires it to still not mention `projectKey` — so the day upstream admits that key, the guard goes red instead of the package quietly continuing to promise a 400. Each port takes its own snapshot, so a consumer that wants one self-consistent answer about a spec that can still change under it should read `spec.unknownKeys` off the reader — which comes from the same snapshot as the four slots — and send that materialized data rather than the live object. |
 | `scripts/run-approvals-feed-unknown-test.mjs` | The approvals feed tells three states apart: **N items waiting**, **nothing waiting**, and **this fetch did not come back, so we do not know**. Every way a fetch can fail (the call throwing or rejecting, a body that is not an object, a `livePending` section that is not an array — including the `null` seen in the field, a `pending` that is not an array or holds a malformed row) publishes `{kind:'unknown', why, at, mode}` on the subscription — never an empty snapshot and never silence. Real snapshots carry `kind:'snapshot'`; `snapshot()` still answers only with the last real one (a fact about the past) while `reading()` answers whether it is current (`unobserved` / `present` / `unknown`). Recovery always publishes a real snapshot again, even when the contents are byte-identical to before the failure. An unknown reading is never counted as zero: the awaiting-decision counts read `null`, the view is empty, and the tracker reports no removals, so cards on screen are not retracted for a failed fetch. Retry, backoff and circuit-breaking are unchanged. |
 | `scripts/run-approval-resolution-test.mjs` | The single discriminated union for **how an approval decision ended** (`ApprovalResolution`: `decided` / `not_sent` / `unsettled`) and its one mapping entry `approvalResolutionOf`: every outcome of the durable-park leg (12 shapes) and of the in-stream frame / suspended-ask leg (4 shapes) lands on exactly one arm and cause; the three meanings of `decision: 'unresolved'` (retracted card, refused edit, respond that never settled) land on three different arms, with `retracted` winning when both flags are set; an interrupted durable card really posts a deny, so it is `unsettled` (`interrupted`), never `not_sent`; a safety stop never claims the decision left the package, and a refusal is only attributed to the engine when the outcome carries positive evidence (a wire error code, or the pointer key the engine mints on a rejection body) — an aborted or code-less decide failure is reported as a plain decide failure; the decision word is passed through without re-validating the closed set; an unreadable outcome is `unsettled` (`unreadable`), never guessed as `decided`; both cause vocabularies are frozen tuples with every word covered by a case, plus the three predicates; the approval-outcome note (`approvalOutcomeNoteOf`) is now derived from the union and compared key-by-key against a reference copy of its previous logic over the released inputs, with a self-check that the comparison can fail; a source-text pin asserts every `return` carrying `respondRefusal` also carries `'unresolved'`. No behaviour change: the existing outcome types and keys are untouched. From 0.80.0 that last pin reads the syntax tree instead of scanning lines: the same return had been rewritten across several lines with conditional spreads, a shape a line-wise search misses entirely, which would have quietly turned the pin into a check of nothing. From 0.83.0 the parked leg's card-stage failure carrying `editRefused` (an edited approval on a card whose tool arguments were not available) reads as not sent with the `edit_refused` cause, the same cause the live leg already had; the flag counts only as a strict `true` on the card stage. |
-| `scripts/run-panel-cycle-identity-test.mjs` | The **cycle identity** on agent-panel events and the fleet ledger's **departure read-out**: a background agent may be revived under the same id, so `fleet-row` and `end` events now carry the wire's own `cycleSeq` / `startedAt` when present (absent means the row has no notion of generations, never "generation one"), `isStaleEngineAgentPanelEnd` is the single rule for ignoring a late `end` from a previous cycle (only when both sides carry a comparable identity; absence never drops a real terminal), a changed `cycleSeq` is a new cycle for usage stickiness and buffer coalescing, the notification lane carries `cycleSeq` only when the wire really sent `seq`, and `task_remove` frames reach the host through `onTaskRemoved` with `removeReason` / `cycleSeq` verbatim, a stale previous-generation removal leaving the newer row in place. A terminal row held back because the consumer has no such row yet is also released by the keys it carries itself (its transcript id, or the delegating call id of a subagent already on screen under its wire id), since the key tables are only written once a row has actually been published — the release still goes through the one funnel, so the subagent stays one row; a running frame arriving after such a held terminal row is a stale snapshot when both sides carry a comparable generation and it matches (no event, the held row keeps its final usage), a revival when the frame is provably newer (the held row is dropped), and is treated as a revival when neither side can be compared. A subagent lifecycle event carries `agentType` only from an honest source — the fleet row's own agent type, recorded before it is folded into the row label — and omits the key when there is none, never substituting the display name |
+| `scripts/run-panel-cycle-identity-test.mjs` | The **cycle identity** on agent-panel events and the fleet ledger's **departure read-out**: a background agent may be revived under the same id, so `fleet-row` and `end` events now carry the wire's own `cycleSeq` / `startedAt` when present (absent means the row has no notion of generations, never "generation one"), `isStaleEngineAgentPanelEnd` is the single rule for ignoring a late `end` from a previous cycle (only when both sides carry a comparable identity; absence never drops a real terminal), a changed `cycleSeq` is a new cycle for usage stickiness and buffer coalescing, the notification lane carries `cycleSeq` only when the wire really sent `seq`, and `task_remove` frames reach the host through `onTaskRemoved` with `removeReason` / `cycleSeq` verbatim, a stale previous-generation removal leaving the newer row in place. A terminal row held back because the consumer has no such row yet is also released by the keys it carries itself (its transcript id, or the delegating call id of a subagent already on screen under its wire id), since the key tables are only written once a row has actually been published — the release still goes through the one funnel, so the subagent stays one row; a running frame arriving after such a held terminal row is a stale snapshot when both sides carry a comparable generation and it matches (no event, the held row keeps its final usage), a revival when the frame is provably newer (the held row is dropped), and is treated as a revival when neither side can be compared. A subagent lifecycle event carries `agentType` only from an honest source — the fleet row's own agent type, recorded before it is folded into the row label — and omits the key when there is none, never substituting the display name Since 0.95.0 the fleet ledger also ends the display of a stopped sub-agent early: when a row that the row-frame lane settled as killed leaves with a settled-level removal (no reason word on the frame), it is dropped from the retention window at once. The rule only shortens the display of a row that is already terminal — it never settles anything and never touches the terminal facts the detail view reads. A statement-level removal, any other terminal word, a late removal for an older cycle, and a row only inferred from a notification are all left alone, a removal that arrives before its terminal frame leaves no trace, and a by-id removal call (per session) follows the same single implementation. |
 | `scripts/run-workflow-size-warning-test.mjs` | The **workflow size warning** verdict shared by every host footer / panel: a three-state result (`warn` / `ok` / `unknown`) read off the optional fleet view keys, where an unknown size is never reported as a normal one (absent `totalCount` / `tokens` without positive over-cap evidence is `unknown`, naming the missing keys), positive evidence on either axis wins regardless of absent keys, the per-agent denominator uses the engine's started count only when it is not below done+failed (a smaller value is a stale reading), otherwise falls back to the done+failed lower bound only when both keys are present — and a lower-bound denominator only yields an upper bound of the projection, which can prove *within cap* (`ok`, flagged) but never *over cap* (`unknown`, with the upper bound exposed) — and the prior is used only when the engine itself reports zero started agents; cap precedence env > explicit guideline > default, prototype keys never act as a guideline, the env reader is pure and does not fall through to the second name on a bad first value; caps, guideline table, env names and the three copy variants are single-sourced |
 | `scripts/run-approvals-stream-live-capability-test.mjs` | The engine's live-approval-push self-description (`capabilities.approvalsStreamLive`, engine ≥7.87.1), read the same four-state way as its four sibling capability readers: an absent key is reported as not reported (never folded into `false`), the value must be a strict boolean, and the one decision the feed consumer needs — whether it must keep pulling suspended asks itself — is answered by `livePendingNeedsReconcile`, which only says no when the engine explicitly says it pushes. |
-| `scripts/run-model-facing-anchors-test.mjs` | Five places where model-facing engine text is parsed back into card slots: the task-output cursor semantics machine-readable key wins over the spool marker text (and its absence is never read as `full`), the `--- result (…) ---` line accepts a parenthesised note, the bash header accepts the engine's exit-note and image variants and carries the note into the same benign-exit slot the structured path uses, the notification seed scan treats both producers' terminal words as terminal, and the parked background-task phrase is recognised. |
-| `scripts/run-execution-lane-capability-test.mjs` | The deployment execution-lane self-description (`capabilities.executionLane`, engine ≥7.86.0), read the same four-state way as its three sibling capability readers. An absent key is reported as "not reported" and never as "tools do not run on this host" as a fact about the deployment; since every server at or above the support floor reports the key, the single "do tools run here" predicate answers `false` for a body that reported nothing (from 0.92.0, so no skill `baseDir` is sent on a reading that proves nothing), and keeps using the caller's own inference only while nothing has been observed yet. The implication is one-way (`toolsOnThisHost: false` means `baseDir` is never sent; `true` is only a necessary condition), `provider` is read as an open non-empty string rather than a hand-copied closed set and is sanitised and bounded before display, `toolsOnThisHost` must be a strict boolean, extra members never cross, the tee never throws, stale generations are dropped whole, and the single "do tools run here" predicate uses the bit when it is present. Current engines (≥7.106.0) also say whether the lane's tools sit behind an OS isolation boundary (`isolated`, present only when the lane has tools of its own); the reading carries it when it is a strict boolean and otherwise records which of three absences it is — a lane without tools on a current engine, an older engine, or a value that cannot be read (including a current engine that says tools run on this host yet omits the flag it must then send) — none of which is ever read as `false`, and an unreadable value never costs the rest of the lane reading. The single "no OS sandbox on this machine" predicate (`isolated` false and tools on this host) answers yes, no or unknown under three-valued logic, one sentence goes with the yes answer, the operator diagnostics segment is read by the same rule, and the engine's own projection for every lane is replayed: only the host lane answers yes |
-| `scripts/run-live-pending-ask-test.mjs` | Suspended in-stream asks (the `livePending` section of `GET /v1/approvals`): an absent key reads as not reported (never an empty list or a zero count), an unreadable section leaves the previous snapshot in place, rows are narrow-read (malformed rows dropped and counted, optional flags only-if-true, extra members never cross), the feed digest covers the section so an ask appearing or settling produces a snapshot, the host-gated reconcile cadence re-lists while the push leg is connected and backs off when nothing changes, the tracker surfaces each `approvalId` once and reports it gone once, and the decision runs through the same respond chain as the in-stream frame leg — with the card stating that the tool arguments are not visible and edits on that card never forwarded. Since 0.73.4 a push-side fetch that did not get a readable body (thrown, or an unreadable `livePending` section) is owed a look: it is retried with doubling back-off until one fetch succeeds (never abandoned by count), cancelled by any successful fetch, cleared on stop, and switchable off. |
+| `scripts/run-model-facing-anchors-test.mjs` | Five places where model-facing engine text is parsed back into card slots: the task-output cursor semantics machine-readable key wins over the spool marker text (and its absence is never read as `full`), the `--- result (…) ---` line accepts a parenthesised note and reads it in linear time even on crafted input (a note can no longer swallow the start of the next line), the bash header accepts the engine's exit-note and image variants and carries the note into the same benign-exit slot the structured path uses, the notification seed scan treats both producers' terminal words as terminal, and the parked background-task phrase is recognised. |
+| `scripts/run-execution-lane-capability-test.mjs` | The deployment execution-lane self-description (`capabilities.executionLane`, engine ≥7.86.0), read the same four-state way as its three sibling capability readers. An absent key is reported as "not reported" and never as "tools do not run on this host" as a fact about the deployment; since every server at or above the support floor reports the key, the single "do tools run here" predicate answers `false` for a body that reported nothing (from 0.92.0, so no skill `baseDir` is sent on a reading that proves nothing), and keeps using the caller's own inference only while nothing has been observed yet. The implication is one-way (`toolsOnThisHost: false` means `baseDir` is never sent; `true` is only a necessary condition), `provider` is read as an open non-empty string rather than a hand-copied closed set and is sanitised and bounded before display, `toolsOnThisHost` must be a strict boolean, extra members never cross, the retired tee port is a no-op that never throws, the reading comes only from the capabilities cache's landed response, and the single "do tools run here" predicate uses the bit when it is present. Current engines (≥7.106.0) also say whether the lane's tools sit behind an OS isolation boundary (`isolated`, present only when the lane has tools of its own); the reading carries it when it is a strict boolean and otherwise records which of three absences it is — a lane without tools on a current engine, an older engine, or a value that cannot be read (including a current engine that says tools run on this host yet omits the flag it must then send) — none of which is ever read as `false`, and an unreadable value never costs the rest of the lane reading. The single "no OS sandbox on this machine" predicate (`isolated` false and tools on this host) answers yes, no or unknown under three-valued logic, one sentence goes with the yes answer, the operator diagnostics segment is read by the same rule, and the engine's own projection for every lane is replayed: only the host lane answers yes |
+| `scripts/run-live-pending-ask-test.mjs` | Suspended in-stream asks (the `livePending` section of `GET /v1/approvals`): an absent key reads as not reported (never an empty list or a zero count), an unreadable section leaves the previous snapshot in place, rows are narrow-read (malformed rows dropped and counted, optional flags only-if-true, extra members never cross), the feed digest covers the section so an ask appearing or settling produces a snapshot, the host-gated reconcile cadence re-lists while the push leg is connected and backs off when nothing changes, the tracker surfaces each `approvalId` once and reports it gone once, and the decision runs through the same respond chain as the in-stream frame leg — with the card stating that the tool arguments are not visible and edits on that card never forwarded. Since 0.73.4 a push-side fetch that did not get a readable body (thrown, or an unreadable `livePending` section) is owed a look: it is retried with doubling back-off until one fetch succeeds (never abandoned by count), cancelled by any successful fetch, cleared on stop, and switchable off. The session scope of the suspended-ask view, count and tracker must be chosen explicitly (one session, all sessions, or identity pending): an unresolved or malformed scope answers null counts, an empty view and a tracker that never emits, and the three entry points refuse to compile without a scope. |
 | `scripts/run-leader-conflict-test.mjs` | The leader-run terminal `needs_human` + `result.conflict` (engine ≥7.83.0), read once for all three shells. A `result` without a `conflict` key is reported as `none` and worded as "no conflict details" rather than "no conflict": on the wire it is indistinguishable from an older engine that never reports one, and the reader does not pick a side. A `conflict` that is present but unreadable is a third word, never folded into `none`, because a tree that carries conflict markers must not be rendered as clean. `filesTruncated` is honoured only as `true` (absent means the list is complete; any other value makes the section unreadable), `workers[].applied` is passed through as the boolean fact it is, `salvaged[].patch` is handed to the save path byte-for-byte (an artifact, not screen text; a malformed row is dropped alone, and a salvage list that is present but cannot be fully read is flagged rather than rendered as "no patches"), and `rejHead` — the one on-screen diagnostic — is escaped and bounded before display. The record never throws on hostile input, extra members do not cross, arrays are fresh copies, and the single sentence minted here names the base, the files, which branches landed and which did not, and how many patches can be saved, without ever suggesting a retry: `needs_human` is a run waiting for a person, and the no-details sentence says plainly that it is not proof of a merge |
 | `scripts/run-mcp-reconnect-test.mjs` | The in-session MCP re-dial verb (`POST /v1/sessions/:id/mcp/reconnect`, engine ≥7.85.0), consumed. The single discriminant is `outcome` and all three answers are HTTP 200, so the reader branches on the word and never on the status; the `unsupported` answer carries exactly five keys and the reader refuses to invent a zero or an empty list for the four fields the engine did not produce, while `accepted` / `refused` treat those four as required and go malformed when one is missing. The tool roster follows the **presence** of `toolNames` (absent = untouched, empty = withdrawn), the connection record passes `errorCode` through as an open set, and every remote-authored string is sanitised and bounded before display. Failures are classified by `errorCode` alone, a missing code is reported as unknown rather than guessed, the verb never throws, and the request-side guard (non-empty name, ≤190 chars) stops a call that the contract would reject anyway. The capability bit reads absent as "cannot tell" rather than "unavailable", and the one sentence the contract insists every UI carries — that re-dialing is a transaction, not a refresh — is minted here once. The engine's answer-deadline reply — a 503 whose machine code says the request stopped waiting before the re-dial settled — is its own verdict arm: it is read by that code alone and never by the status, carries no outcome, and is never rendered as a failure, because the re-dial was not cancelled and may still be running; the one sentence for it is minted here once, and the gate pins that no server release up to 7.117.0 contains the code at all |
 | `scripts/run-core-value-ports-test.mjs` | The port-injection seam for ten **engine value-level** facilities (autonomous-loop prompt assembly, permission-rule loosening, tool-policy composition, protocol/retired-name/grammar lookups, rule compilation, the discussion workflow name). This package cannot re-export them (the engine barrel drags Node built-ins into the browser bundle), so it declares the ports and honest-absence readers; a Node host installs the engine's own functions verbatim. The guard pins: every reader returns `undefined` when nothing is installed (never a fabricated empty array or default policy), arguments and results pass through by reference, engine errors propagate unchanged, partial installs read partially, restore functions unwind to the previous bag, and the module source has zero engine imports From 0.90.0 the mirror of the allow/deny policy options is also compared with the engine's own declaration, both ways: the same set of option keys (the engine added an optional `scope` — whether an allow-list hit is a stated grant or only a bound on which tools a lane may see), with `scope` optional and its two-word union and the invalid-name union spelled identically; the value handed to the installed function is passed through untouched. Since 0.94.0 there are twelve ports: the engine's allow-list reader joins them, with the same absence and pass-through rules. From 0.94.0 the seam also carries the engine's five route-adjudication functions (`routeKeyOf`, `joinRequestUrl`, `resolveRouteCredential`, `adjudicateModelRoute`, `routePairingStatus`): a host that builds the full `CoreValuePorts` bag fails to compile until it adds them (the gate compiles that bag in memory and checks each missing member is named), the engine's own functions type-check against the mirrors, and each reader answers `undefined` while its port is not installed. |
 | `scripts/run-read-face-posture-projection-test.mjs` | The operator-face `readFace: ReadFacePosture` reader (server >=7.65.0). Three ways of "can't say" are pinned to three different, literal sentences, and none of them may read as "nothing is pinned" — that statement belongs to exactly one case, `face: null`, which is a positive fact reported by the engine, not an absence: not having read an operator response yet, having read one from an engine too old to report the key, and the engine actually saying nothing is pinned are three different next steps for an operator and must not collapse into each other. `source` is read as an open set (the server's closed four words plus an escape hatch) rather than narrowed to an enum, so a new word added upstream is not silently turned into a bad reading. The free-text `note` is sanitized and length-capped before it is ever rendered. A companion pure function flags disagreement between this face and the tenant-facing `capabilities.readFace` — silent only when the two actually agree, honest-absent when either side cannot be read at all, never asserting agreement as a fact. The gate's last leg reads the installed SDK's own `openapi.yaml` directly rather than restating the schema in prose, so the package's leniency cannot quietly drift from the real contract |
-| `scripts/run-display-body-test.mjs` | The engine wraps text it hands a model in a fence — an opening marker naming the payload, the payload itself, and a closing marker — so the model reads it as data and not as instructions. That fence is minted and read in one place here, which makes stripping it for a human reader this package's job rather than each shell's: a shell that renders the envelope verbatim is showing a person a defence that was written for a model. The reader answers with a discriminated union — fenced, with the label and the payload, or not fenced, with the text as it came in — and it reaches that answer through the **same** matcher the mint side registers, never a second copy of it; the guard proves that by walking the syntax tree of every source file and requiring exactly one literal carrying the marker text, and by requiring the reader's own body to contain no matcher of its own. Eighteen shapes are run through both entry points and required to agree line for line. Anything the package does not recognise — a near-miss in the wording, a hyphen where the marker has a dash, a different case, an opening marker with no close, a close before an open, a truncated close, or any non-whitespace byte outside the pair — comes back unfenced with the input returned **verbatim**: no guessing, no trimming, no repair, because a half-stripped envelope puts a sentence on screen that nobody wrote. Only the outermost layer is removed, so a nested fence, or one forged inside the payload, survives byte-for-byte in the body — those bytes are part of what the engine said, not part of this protocol. Nothing else is washed: control characters, leading and trailing whitespace and a twenty-thousand-character payload all pass through untouched, and so does the label, because sanitising and length-capping belong to the mint point that puts a string on a screen and a passage of text must not have two launderers. A value that is not text is answered with **nothing at all** rather than with an empty payload: the reader never stringifies it, never calls its `toString`, and never emits `[object Object]`, and it does not hand back a body of zero length either — an empty payload is a real reading (a fence can legitimately wrap nothing, and an empty string is an empty string), so folding "there was no readable text" into it would leave a caller unable to show a degraded line at all. Those three stay apart: no text yields nothing, an empty string yields an unfenced empty payload, and an empty fenced payload yields a fenced one with its label. The `fenced` discriminator is always present on a reading, and the label key exists only on the fenced arm, so a missing label is never rendered as an empty one. One shape needed more than the whole-string match this started with. When the engine reports back from a delegated run, the fence is only **one section** of the report: ahead of it sit a frame header, a handful of optional field lines and a section label, behind it a closing instruction addressed to the model, an optional internal identifier and a usage block. A matcher anchored to both ends of the input answers *not fenced* on that, and the whole scaffold - written for a model - goes on screen. The reader therefore also locates the fence **inside** a recognised report frame, using four anchors that are always present and always byte-for-byte fixed, and hands the located slice back to the same single matcher rather than a second one; the guard assembles its corpus from the engine package actually installed (the fence from that package's own constructor, the frame lines read structurally out of the minting file and then compared byte-for-byte with what this package registers), so a rewording or a reordering upstream turns the guard red the day it lands. Three readings are pinned one cell each: a fenced result section yields the payload the delegate actually wrote; a partial-findings section yields that text and says which of the two it is, with the prefix line kept out of the payload; and a section the engine filled with its own no-text sentinel yields an empty payload with a reason, which stays distinguishable from an input that was simply an empty string. Whatever surrounded the fence is returned alongside rather than dropped - the bytes before it, the slice itself and the bytes after it reassemble into the input exactly - and the payload never contains a line of the frame. The criterion deliberately does **not** enumerate the lines outside the fence: several of those field lines carry interpolated untrusted text and cannot be told apart from prose, so requiring every one of them to be recognised would mean that a single new field line upstream sends every failed report back to being unreadable, and a failed report is exactly when a person most needs to read what the delegate said. Fourteen negative shapes hold the line against the easy widening, strip anything that looks like a fence: a fence sitting in ordinary text, a missing frame header, a different sentence where the closing instruction belongs, a report with no closing instruction at all while the fence sits at the very end, a header and label in the wrong order, mismatched open and close tags, an open with no close, a bare unfenced section, a stray line between the label and the fence or between the fence and the closing instruction, an entirely absent section, a no-text sentinel with another line after it, a partial-findings prefix followed by something that is not a fence, and the whole report in carriage-return line endings all come back unfenced with the input verbatim and no frame at all. Above all, **provenance is not in the text**: locating a fence inside a frame happens only when the caller states where the bytes came from, because four anchors can only recognise a shape and never prove an origin. The default reading is byte-for-byte what it was before, so a passage of ordinary prose that happens to quote a report - with real warnings on either side of the quoted part - is returned untouched and those warnings stay on screen; a caller that does state the origin gets the located reading, and even then every surrounding byte comes back alongside. Twenty-three malformed origin values fall back to the narrower default without throwing: a near-miss in case, a camel-cased spelling, the right word padded with spaces or tabs or a newline or a zero-width character, a string wrapper object, an object whose `toString` or `valueOf` reports the right word, and an object whose converters both throw. The guard compares the caller’s argument for **exact equality** and nothing else — no trimming, no stringifying, no calling the value’s own converters, because that would let a value of unknown provenance choose its own lane — and a source-level cell requires that the argument reach the comparison unreassigned and unnormalised, with its own two-way check that those patterns speak. The two accepted words are read off the published type rather than copied into the guard; adding a third word later is a type-compatibility change for any caller that switches exhaustively on them. Each of the three anchors must also be **unique** in the text, and ambiguity means the reader declines. The reason is not hypothetical: the frame header interpolates the task's own description verbatim, and the engine only requires that description to be a string, so it can carry newlines and a complete set of protocol lines. Any rule that picks one candidate out of several can therefore be made to pick the planted one, hiding the real result among the surrounding bytes - a guard cell reproduces exactly that, with a planted section and a real one, and requires the reader to decline and the real text to stay on screen. Two reports back to back are the same ambiguity and are declined the same way, with both payloads left visible; a delegate that quotes any one of the three anchor lines inside its own answer also falls back to the input verbatim, which is the registered cost of the rule, and a discrimination cell shows the same corpus reads cleanly once the quoted line is gone. The two legs are not the same shape either: the forked one ends at its closing instruction with no trailing bytes at all, and that real shape has its own cell. The witness arm reads each leg only inside its own array of lines, decodes every extracted literal to its **runtime** value rather than trusting the spelling in the source, and fails loudly if it cannot - an escape rewrite upstream leaves the runtime label unchanged while the spelling diverges, and since that same extracted value builds the corpus and serves as the expectation, trusting the spelling would close a self-proving loop. The two payload labels are therefore also pinned in the guard and compared against what was extracted, so an equivalent rewrite stays green while a real rename turns red the day it lands. A delegate that quotes the frame lines inside its own answer does not move the location, and those quoted lines survive in the payload byte-for-byte |
+| `scripts/run-display-body-test.mjs` | The engine wraps text it hands a model in a fence — an opening marker naming the payload, the payload itself, and a closing marker — so the model reads it as data and not as instructions. That fence is minted and read in one place here, which makes stripping it for a human reader this package's job rather than each shell's: a shell that renders the envelope verbatim is showing a person a defence that was written for a model. The reader answers with a discriminated union — fenced, with the label and the payload, or not fenced, with the text as it came in — and it reaches that answer through the **same** matcher the mint side registers, never a second copy of it; the guard proves that by walking the syntax tree of every source file and requiring exactly one literal carrying the marker text, and by requiring the reader's own body to contain no matcher of its own. Eighteen shapes are run through both entry points and required to agree line for line. Anything the package does not recognise — a near-miss in the wording, a hyphen where the marker has a dash, a different case, an opening marker with no close, a close before an open, a truncated close, or any non-whitespace byte outside the pair — comes back unfenced with the input returned **verbatim**: no guessing, no trimming, no repair, because a half-stripped envelope puts a sentence on screen that nobody wrote. Only the outermost layer is removed, so a nested fence, or one forged inside the payload, survives byte-for-byte in the body — those bytes are part of what the engine said, not part of this protocol. Nothing else is washed: control characters, leading and trailing whitespace and a twenty-thousand-character payload all pass through untouched, and so does the label, because sanitising and length-capping belong to the mint point that puts a string on a screen and a passage of text must not have two launderers. A value that is not text is answered with **nothing at all** rather than with an empty payload: the reader never stringifies it, never calls its `toString`, and never emits `[object Object]`, and it does not hand back a body of zero length either — an empty payload is a real reading (a fence can legitimately wrap nothing, and an empty string is an empty string), so folding "there was no readable text" into it would leave a caller unable to show a degraded line at all. Those three stay apart: no text yields nothing, an empty string yields an unfenced empty payload, and an empty fenced payload yields a fenced one with its label. The `fenced` discriminator is always present on a reading, and the label key exists only on the fenced arm, so a missing label is never rendered as an empty one. One shape needed more than the whole-string match this started with. When the engine reports back from a delegated run, the fence is only **one section** of the report: ahead of it sit a frame header, a handful of optional field lines and a section label, behind it a closing instruction addressed to the model, an optional internal identifier and a usage block. A matcher anchored to both ends of the input answers *not fenced* on that, and the whole scaffold - written for a model - goes on screen. The reader therefore also locates the fence **inside** a recognised report frame, using four anchors that are always present and always byte-for-byte fixed, and hands the located slice back to the same single matcher rather than a second one; the guard assembles its corpus from the engine package actually installed (the fence from that package's own constructor, the frame lines read structurally out of the minting file and then compared byte-for-byte with what this package registers), so a rewording or a reordering upstream turns the guard red the day it lands. Three readings are pinned one cell each: a fenced result section yields the payload the delegate actually wrote; a partial-findings section yields that text and says which of the two it is, with the prefix line kept out of the payload; and a section the engine filled with its own no-text sentinel yields an empty payload with a reason, which stays distinguishable from an input that was simply an empty string. Whatever surrounded the fence is returned alongside rather than dropped - the bytes before it, the slice itself and the bytes after it reassemble into the input exactly - and the payload never contains a line of the frame. The criterion deliberately does **not** enumerate the lines outside the fence: several of those field lines carry interpolated untrusted text and cannot be told apart from prose, so requiring every one of them to be recognised would mean that a single new field line upstream sends every failed report back to being unreadable, and a failed report is exactly when a person most needs to read what the delegate said. Fourteen negative shapes hold the line against the easy widening, strip anything that looks like a fence: a fence sitting in ordinary text, a missing frame header, a different sentence where the closing instruction belongs, a report with no closing instruction at all while the fence sits at the very end, a header and label in the wrong order, mismatched open and close tags, an open with no close, a bare unfenced section, a stray line between the label and the fence or between the fence and the closing instruction, an entirely absent section, a no-text sentinel with another line after it, a partial-findings prefix followed by something that is not a fence, and the whole report in carriage-return line endings all come back unfenced with the input verbatim and no frame at all. Above all, **provenance is not in the text**: locating a fence inside a frame happens only when the caller states where the bytes came from, because four anchors can only recognise a shape and never prove an origin. The default reading is byte-for-byte what it was before, so a passage of ordinary prose that happens to quote a report - with real warnings on either side of the quoted part - is returned untouched and those warnings stay on screen; a caller that does state the origin gets the located reading, and even then every surrounding byte comes back alongside. Twenty-three malformed origin values fall back to the narrower default without throwing: a near-miss in case, a camel-cased spelling, the right word padded with spaces or tabs or a newline or a zero-width character, a string wrapper object, an object whose `toString` or `valueOf` reports the right word, and an object whose converters both throw. The guard compares the caller’s argument for **exact equality** and nothing else — no trimming, no stringifying, no calling the value’s own converters, because that would let a value of unknown provenance choose its own lane — and a source-level cell requires that the argument reach the comparison unreassigned and unnormalised, with its own two-way check that those patterns speak. The two accepted words are read off the published type rather than copied into the guard; adding a third word later is a type-compatibility change for any caller that switches exhaustively on them. Each of the three anchors must also be **unique** in the text, and ambiguity means the reader declines. The reason is not hypothetical: the frame header interpolates the task's own description verbatim, and the engine only requires that description to be a string, so it can carry newlines and a complete set of protocol lines. Any rule that picks one candidate out of several can therefore be made to pick the planted one, hiding the real result among the surrounding bytes - a guard cell reproduces exactly that, with a planted section and a real one, and requires the reader to decline and the real text to stay on screen. Two reports back to back are the same ambiguity and are declined the same way, with both payloads left visible; a delegate that quotes any one of the three anchor lines inside its own answer also falls back to the input verbatim, which is the registered cost of the rule, and a discrimination cell shows the same corpus reads cleanly once the quoted line is gone. The two legs are not the same shape either: the forked one ends at its closing instruction with no trailing bytes at all, and that real shape has its own cell. The witness arm reads each leg only inside its own array of lines, decodes every extracted literal to its **runtime** value rather than trusting the spelling in the source, and fails loudly if it cannot - an escape rewrite upstream leaves the runtime label unchanged while the spelling diverges, and since that same extracted value builds the corpus and serves as the expectation, trusting the spelling would close a self-proving loop. The two payload labels are therefore also pinned in the guard and compared against what was extracted, so an equivalent rewrite stays green while a real rename turns red the day it lands. A delegate that quotes the frame lines inside its own answer does not move the location, and those quoted lines survive in the payload byte-for-byte. For a recognised report frame the reader also hands back the header slots a shell used to assemble by counting lines: the state line (read only when it sits immediately after the frame header and is the only state line in the frame, so a line planted in the task description cannot masquerade as the engine's own), an error line and a usage line (each only when it is the only one of its kind), and every line it did not claim, verbatim and in order; when the state line cannot be read from that position all four slots stay absent and every byte stays in the remainder |
 | `scripts/run-display-cap-order-test.mjs` | The order in which untrusted text is sanitised and length-capped, across every mint point that puts an engine- or database-supplied string on a screen. The sanitiser rewrites each invisible character as a six-character escape, so capping the **raw** string first and escaping afterwards hands the screen six times the width that was budgeted — a forty-character allowance becomes two hundred and forty. The guard does not hardcode that allowance, because each mint point wraps its field in different fixed prose and the prose moves: it anchors on the deciding quantity instead, feeding one benign and one control-character input of the same length through the same mint and requiring the second not to come out longer. That criterion is immune to wording changes and stays sensitive to the expansion, and it is `<=` rather than `==` on purpose — a correct escape-then-cap backs the cut off a partially-consumed escape token, so the control-character line is legitimately the shorter of the two, and demanding equality would score that avoidance as a regression. Each mint is bracketed by two positive controls (the input really reaches the screen; the cap really engages) and the expansion predicate is shown to turn red against a deliberately cap-then-escape reference, so an all-green run cannot mean the guard simply measured nothing. The shared mint point is checked directly for the two avoidances it owes — never splitting an escape token in half, which would leave something on screen that looks like the beginning of a complete answer, and never splitting a legal surrogate pair, which would manufacture the very lone surrogate the sanitiser exists to catch |
 | `scripts/run-seat-task-request-origin-test.mjs` | Where every field of the seat lane's send-message payload comes from, and whether it actually lands anywhere. The seat payload is a closed interface this package mints itself, and most of its fields are meant to ride verbatim onto the engine's request body — two facts nothing used to connect, so both directions could drift in silence. A seat field could be named after a request position that does not exist, in which case a client writes to it, the wire carries it, the engine ignores the whole key, and the screen shows a switch that does nothing; conversely a new request position could arrive with no seat to sit in, which is **structural** absence — the closed set *is* the carrier, so a decision missing from it has nowhere to be put at all, the same shape logged when the effort dial had no seat. The guard turns each field's origin into data: either it names the request position it forwards to, or it is declared seat-local with a written reason, and the two are mutually exclusive. Forwarding claims are then checked against the **installed** SDK's type declarations, parsed rather than restated — a hand-copied list of position names would only ever prove that two transcriptions agree. The parser is held to reading top-level positions only, since a nested option object's inner keys would otherwise be mistaken for positions of the request itself, and it proves that discrimination on synthetic input before any verdict is given. The two subagent fields carry a standing regression pin, and the retention window's inner keys are read from the declaration the same way, so a seat that offers a tunable window cannot offer one the wire has no room for |
 | `scripts/run-wire-auth-source-test.mjs` | **When** the outbound credential is read. A literal string is consumed at construction — the transport captures it in a closure and every later request reuses that one copy — so once the engine is replaced by another session and the credential rotates, a long-lived client keeps presenting the old one and the only way out is to rebuild the client along with everything hanging off it. The credential position now also accepts a getter that is called **once per outbound request**. The guard anchors on the deciding quantity, which is not "was the getter called" — reading once at construction and reusing the result would satisfy that too, and is exactly the shape being removed — but *which read produced the value on the wire*: it changes the getter's answer between two requests through the same client and requires the second request to carry the new one, and it requires construction to read the getter **zero** times. The three-state credential semantics are replayed per request rather than assumed: on loopback an unavailable credential sends **no** authorization header at all rather than a fabricated one, off loopback it sends the fail-closed anonymous identity so the deployment answers with an honest 401, and the guard shows a single client moving between those states across successive requests. A getter that throws is fail-soft — the request still goes out under the no-credential branch, because a broken credential port should not take the whole wire down, and the exception may itself carry credential material. The same-origin relay form is checked to stay out of the getter path entirely, and every request is checked to keep the credential in the authorization header only — never in the URL, never in another header. The same getter form is also accepted by two further entry points, and each is judged by its end result. For the workflow activity ledger, connection identity now follows the credential's *source* rather than the value read from it: a string is compared by value, a getter by reference, the same-origin relay declaration by its mode, and a change of form counts as a new source. The guard rotates the getter's answer between two warm-up calls and requires that no second stream opens and that the ledger keeps every frame it had already collected — before, a host could only pass a freshly read string, so each rotation looked like a different connection and the ledger was replaced by an empty one. A reconnect of that same ledger and the monitor's detail reads must then carry the rotated value, and the getter must be read exactly as many times as requests go out, so comparing identities never reads the credential. As the negative control, a different getter returning the same value opens its own stream, replaces the ledger, and the monitor holding the previous getter stands down instead of taking the slot back. For the detach cancel fallback, arming with a getter reads it zero times; the pick-up the host calls on its signal path reads it at that moment and still hands out only the two established forms, so an unchanged copy of the host's cancel leg sends the credential that is current at send time. The three-state semantics match the wire client (no authorization header on loopback when nothing is available, the anonymous identity elsewhere), a throwing getter never makes the pick-up throw, and string or loopback arms are handed back as the very same object. The widened inputs are checked with the compiler: the getter form is assignable, every previous form still is, the pick-up still narrows to the two forms, and illegal forms are rejected |
-| `scripts/run-subagent-durable-divert-test.mjs` | The side-channel that keeps a **sub-agent's** content out of the leader's transcript, on the replay leg. A content frame stamped with a parent tool-call id belongs to a child, and rendering a child's tokens as the leader's own text is the pollution this divert exists to prevent — but the predicate only listed the four **live** frame shapes, while the durable leg replays the same segment in its **aggregated** form. Those frames fell straight through onto the main projection path, which is how a reconnect or a resumed session ended up with the child's answer printed as the leader's. The anchor is unchanged and shared: the parent tool-call id is what says whose frame this is, and whether the frame is an increment or a whole segment has nothing to do with whose it is — judging the two shapes separately is exactly how one of them got missed. Folding the aggregate into a synthetic increment would have been the smaller diff and the wrong one: an increment means *append*, so a segment that already streamed live and then replays whole would be counted **twice**. The two are kept distinct and the aggregate absorbs instead — a whole segment whose prefix is what the buffer already holds replaces it, which also makes a redelivery of the same frame idempotent, and a prefix that does not match falls back to appending both rather than deciding on the engine's behalf which version counts. Segment boundaries stay with the tool frames rather than moving into the aggregate arm, since closing there would turn a second replay of one segment into a second entry, and the increment arm is pinned to keep appending so a token run that happens to be a prefix of the next does not silently lose characters. When the host declares the non-interactive lane, a sub-agent's tool calls and results are also forwarded into the main output with their parent tool-use id (the sub-agent's text and thinking still stay out, as in the reference CLI); without that declaration the output is unchanged. |
+| `scripts/run-subagent-durable-divert-test.mjs` | The side-channel that keeps a **sub-agent's** content out of the leader's transcript, on the replay leg. A content frame stamped with a parent tool-call id belongs to a child, and rendering a child's tokens as the leader's own text is the pollution this divert exists to prevent — but the predicate only listed the four **live** frame shapes, while the durable leg replays the same segment in its **aggregated** form. Those frames fell straight through onto the main projection path, which is how a reconnect or a resumed session ended up with the child's answer printed as the leader's. The anchor is unchanged and shared: the parent tool-call id is what says whose frame this is, and whether the frame is an increment or a whole segment has nothing to do with whose it is — judging the two shapes separately is exactly how one of them got missed. Folding the aggregate into a synthetic increment would have been the smaller diff and the wrong one: an increment means *append*, so a segment that already streamed live and then replays whole would be counted **twice**. The two are kept distinct and the aggregate absorbs instead — a whole segment whose prefix is what the buffer already holds replaces it, which also makes a redelivery of the same frame idempotent, and a prefix that does not match falls back to appending both rather than deciding on the engine's behalf which version counts. Segment boundaries stay with the tool frames rather than moving into the aggregate arm, since closing there would turn a second replay of one segment into a second entry, and the increment arm is pinned to keep appending so a token run that happens to be a prefix of the next does not silently lose characters. When the host declares the non-interactive lane, a sub-agent's tool calls and results are also forwarded into the main output with their parent tool-use id (the sub-agent's text and thinking still stay out, as in the reference CLI); without that declaration the output is unchanged. On that same non-interactive lane the tool-call blocks of the main output carry only the reference shape (type, id, name, input): the display name and the two identity keys the interactive bridge strips before rendering are removed through the very function the bridge uses, the input object itself is left untouched, and the interactive lanes keep the three keys. |
 | `scripts/run-subagent-content-budget-test.mjs` | The **byte** budget on the sub-agent transcript ledger. It used to be bounded only by *counts* — so many entries per child, so many children — and a count is not a budget when a single entry has no ceiling of its own: one tool result carrying an inlined attachment, or one long model answer, and a single slot sits on tens of megabytes. The guard anchors on how many bytes are **still held** after over-filling, not on whether truncation fired, because an implementation that flags the overflow without actually dropping anything satisfies the second and not the first. Dropping is required to leave a record — how much went and where the retained content now starts — and that record has to reach the render plan, because content that vanishes with no marker gives the reader a transcript shorter than what happened with nothing to say so; the record is one per child, updated in place, pinned to the front, and excluded from the budget it describes. Order matters and is checked: oldest entries go first and the live tail is trimmed only as a last resort, since taking the text the user is watching stream while older history survives is the wrong end. The total budget evicts a whole least-recently-used child rather than shaving every child, and the configuration surface is fail-loud on zero, negatives, non-finite and non-integer values — a silently ignored budget is the exact failure this exists to remove — with the rejection proven atomic so a bad second field cannot leave half a configuration behind. The defaults are checked to be a magnitude that can really be reached, since a number too large to hit is a field rather than a budget |
 | `scripts/run-subagent-usage-projection-test.mjs` | Per-subagent usage, split by task. The engine's final accounting carries the delegated spend as **one total** — tokens, turns, task count — and no per-task breakdown, while every sub-flow turn on the stream carries its own usage. This package used to fold that away at the leader/sub-flow divide (a child's output tokens must never reconcile the leader's response length), so a client showing a subagent's detail pane had nothing to print. The split table can therefore only be accumulated from the stream, and this guard pins what that costs. The two existing leader-only arms stay **byte-for-byte unchanged** — the new arm is additive and always carries the sub-flow's own lane proof, so a host cannot mistake a child's numbers for the session window. Attribution is by the engine's own originating-task id — deliberately not a second `taskId`, which the event identity does not carry and whose absence would silently collapse every child under one parent call — falling back to the parent call id; a turn that answers neither is dropped rather than filed under an invented row, because merging two children's ledgers is worse than missing one. Cache-read tokens are read from the **engine's own shape** rather than the mirrored one, since the mirror fills that member with zero when the wire omits it and reading it there would erase the difference between *not reported* and *no cache hit*. A turn that reported no usage at all still counts as a turn and still adds its zeros — the numbers are a lower bound, and dropping the round would make the bound less true, so the honesty bit rides on the row instead and is never spelled `false`; such a round still emits its live arm, because the frame that says "this round has no account" is the one a real-time consumer most needs and the easiest one to drop. The same honesty bit also survives a terminal that carries no statistics at all: what the stream observed is unioned with what the final record says, so a run that already reported an unmeasured round cannot come out the other end looking like an exact zero. Finally the table says whether it is **partial**, and that verdict is anchored on the quantity that actually decides it: the engine's own totals. Turn count and row count must both reconcile before the table claims to cover the whole run; anything else — including totals that cannot be read — marks it partial, so the failure direction is always the safe one (a complete table called partial, never the reverse). The two accounts are kept separate and are never added together or used to correct each other. One more thing the totals cannot settle: the row key has **two namespaces** — the originating-task id and the parent call id it falls back to — and nothing upstream promises they are disjoint, so the same literal can name one child's identity and another child's parent call. Accumulation therefore keys on the origin as well as the id; the delivered table still keys on the bare id, and a cross-namespace clash is merged into one row that says so, with the partial verdict forced, because a row count and a turn count can both reconcile while the attribution behind them is wrong. The table itself is likewise a **per-stream snapshot** handed to the terminal projector by value rather than left on the caller's context: the three terminal projectors are public, so a host may drive one run through the stream and project another's terminal directly on the same context, and a table left behind would be attributed to whoever projects next — silently called complete whenever that run's own totals happen to match. Without a snapshot, both table keys are simply absent |
 | `scripts/run-result-text-backfill-test.mjs` | What happens when the terminal frame's answer text and the text already on screen do not match. A turn's answer normally streams in and the terminal frame carries the same words again, so the two agree — but when the connection drops mid-answer and the reconnect brings the finished version, "this turn already produced assistant text" is true, the terminal fallback is skipped entirely, and the screen stays permanently short of whatever arrived while the stream was down, with nothing to say so. Four cases are pinned. Nothing on screen yet: render the terminal text whole, byte for byte the previous behaviour. On-screen text is a **prefix** of the terminal text: emit only the missing tail, and the guard measures the deciding quantity — the total bytes that reached the screen must equal the terminal text, which fails both for a missing tail and for a re-render that would print the first half twice; when the two are already equal, nothing is emitted at all. Terminal text is a prefix of what is on screen (an engine-side trim): touch nothing, since there is nothing missing and overwriting with the shorter version would erase what the reader already saw. Neither is a prefix of the other: emit **nothing** and raise a fact instead — which version counts is the engine's to say, and appending the terminal version after the streamed one composes a passage nobody ever wrote. That fact carries lengths rather than text, so a renderer is not handed a third version to choose from, and its declared duty is to *reword* the transcript line, never to render more. A cross-segment case proves the comparison reads the whole committed answer rather than the last segment, and the whole thing is driven through the real two-stage path rather than hand-built messages. The comparison only holds if both sides are the same kind of thing — one assistant message against one assistant message — and that depends on the client knowing where a message ends. A tool card is one place a message ends, but not the only one: when a model round finishes and the next one starts writing prose straight away, with no tool call in between, the two passages belong to two different assistant messages even though nothing visible separates them on the wire. Reading them as one used to glue the two passages into a single transcript line, running the end of one sentence into the start of the next, and then handed the terminal comparison a concatenation to check against the engine's **last** message — so a perfectly ordinary multi-part answer was reported to the reader as *the final answer does not match the text streamed above*. The end of a model round is therefore treated as the end of an assistant message: the pending prose is committed and a new message begins. That holds even when the round reports no spend at all, because *this round is over* and *this is what it cost* are two different facts and only the first one decides a boundary. A round belonging to a **sub-flow** decides nothing for the leader, and the test pins all three identity members, empty strings included, against a control that proves the same shape really does divide when no identity is present. The rest of the section is regression: a boundary landing immediately after a tool card must not lose the fallback that lets a card-ends-the-turn answer compare against the prose before the card; repeated boundaries with no prose between them must not mint empty messages or lose track of which message the comparison should read; a reconnect that brings the finished answer still backfills only the missing tail; the leg that already carries whole messages is not divided twice; and a segment boundary that arrives **after** the round ended still replaces the segment authoritatively and hands back the row attribution, without minting a second copy of the passage. A last group covers where a message boundary meets a segment that the engine announces late or not at all, and it pins only the half that is unambiguous. A round whose prose the engine never announces, followed by one it does, used to have the first passage **silently overwritten** by the second one's authoritative text — no transcript line for it and an empty attribution list, so a host had nothing to correct; it now keeps its own line and the attribution is handed over in full, with the offset measured on that passage rather than on the authoritative text. Which of the two passages survives a compliant host's rewrite is deliberately **not** asserted: the row named belongs to an earlier message, exactly as it already did at a tool-call boundary, and the underlying cause — the client splitting messages on its own boundaries while the engine announces segments on content blocks — is recorded as a known limit rather than pinned as a desired outcome. Where the second round's prose arrives as a whole block instead, there is no segment announcement at all and both passages keep their own line **in the order they happened** — previously the whole block was written first and the earlier passage only landed at the close, so the reader saw them reversed. An announcement that arrives after its round has already ended, and whose authoritative text merely extends what was shown, is pinned on the two things that are not in question: the bytes on screen add up to the authoritative text exactly once, and the divergence signal with its offset is still emitted so a host can reword. That ordering is unreachable on the installed engine — the announcement is pushed while the message is still being assembled and the round end only after it is complete, both through one synchronous dispatcher onto one queue — so the case is defensive; the late-announcement path exists because a tool call can be admitted before assembly finishes, which a round end cannot. Finally, one guard names a layer seam rather than a behaviour: the round-end frame is folded into a neutral usage arm that carries **none** of the three identity members, so a round belonging to a forwarded background child arrives with nothing to judge and the sub-flow cutoff cannot reach it. That guard reddening is the signal that identity now survives the fold and the cutoff has become effective |
 | `scripts/run-engine-vocab-floor-test.mjs` | Engine-mirrored vocabularies (structured card whitelist, self-reported tool face, control verbs, recogniser sets) against the *installed* `@sema-agent/core` |
-| `scripts/run-limits-env-failloud-test.mjs` | `SEMA_HEADLESS_*` env-lane limits reject invalid values as loudly as the flag lane (no silent "no budget" runs) |
+| `scripts/run-limits-env-failloud-test.mjs` | `SEMA_HEADLESS_*` env-lane limits reject invalid values as loudly as the flag lane (no silent "no budget" runs) The sentence about whether an attached engine honors an explicit limit flag (`limitsSupportNotice`) and the helper naming the explicitly given limit flags (`explicitLimitFlagNames`) are pinned here: no sentence when the capability probe was not observed or the version is at or above the support floor, a "may ignore" sentence below it, a "cannot be confirmed" sentence when no readable version was reported, and only the three limit flag names are ever echoed. |
 | `scripts/run-streamjson-timing-honesty-test.mjs` | Stream timing & terminal honesty ([2084]): held errored fs-write results release on model progress; a wall-clock stop maps to `error_during_execution` with a truthful salvage note; the synthetic API-error assistant row carries the `<synthetic>` in-message sentinel. Also ([2489], core 5.8.0): the run-limit `errorCode` -> CC subtype map is pinned code by code (`limits.max_{cost,turns,tokens,walltime}_exceeded`), token/wall-clock stops keep the text the engine already produced, and the `failed` event arm shares the one mapping point. The 5.7 dual-vocabulary legs retired with server 6.0.0 (which bundles core 5.8.0); four **retirement negative controls** stand in their place — the retired `status:'timeout'` and the retired codes must fall to the honest fallback subtype and must never drop back to an empty success, so putting any of them back turns the gate red |
 | `scripts/run-background-view-test.mjs` | `createBackgroundView` lifecycle: polling/notify pairing, per-source degrade (`501 → not-configured` vs `unavailable`), the capabilities `scheduler` probe, and dispose really aborting the in-flight fleet snapshot (pure projection lives in the pure suite's W-A segment) |
-| `scripts/run-fleet-view-keys-test.mjs` | The fleet projection views, **both directions**: `FleetTaskView`/`FleetWorkflowView` ⇄ their key lists (compile-pinned) ⇄ what a maximal/minimal row really projects, plus a wire-key coverage ledger (every `FleetTaskRow` key is either projected or carries a written reason why not) and a drift ledger against the shell's render contract. A one-directional assignability check is blind to optional keys — which is how `startedAt` was silently dropped |
+| `scripts/run-fleet-view-keys-test.mjs` | The fleet projection views, **both directions**: `FleetTaskView`/`FleetWorkflowView` ⇄ their key lists (compile-pinned) ⇄ what a maximal/minimal row really projects, plus a wire-key coverage ledger (every `FleetTaskRow` key is either projected or carries a written reason why not) and a drift ledger against the shell's render contract. A one-directional assignability check is blind to optional keys — which is how `startedAt` was silently dropped Two keys that no server ever sends have left the view since 0.95.0 — the queue-depth count on a task row (the SDK still declares it, as a phantom key) and the run-level error code on a workflow row (the fleet workflow row never carries it): whatever the input holds, the view has no such key, and the first is recorded in the not-projected ledger with its reason. Where the shell's render contract still declares them, the drift ledger lists exactly those two as a transitional entry that must equal the measured set, so the day the shell drops its declaration the entry is a zombie and the gate says so. |
 | `scripts/run-usage-verbatim-channel-test.mjs` | The two complementary usage disciplines (core 3.0.0 metering semantics): the CC `ModelUsage` mirror stays pure (five pinned keys, `totalInputTokens` has no seat), while the sema-owned channel forwards the engine `turn_end.usage` object **verbatim** (six keys, incl. `totalInputTokens`) via `last_turn_usage.engineUsage` / `handle.latestEngineUsage` — honest absence on pre-3.0.0 engines, no fabricated zeros |
-| `scripts/run-plan-review-decide-verify-test.mjs` | `decidePlanReview`'s post-decide honesty ([2315]/[2316], engine RB-471 family): a 2xx from the decide endpoint is **not** a terminal — the wire re-pulls the task status and words the outcome by the real shape (still-locked / legal new gate / genuinely left park / unverified), never claiming success it hasn't earned; when the engine answers that the session's stored resume context cannot be read, the outcome names the unreadable row and says the decision was not applied. Driven against a real fake-engine HTTP server through the shipped dist. The outcome queue item also carries a machine-readable `_sema_planReviewOutcome` (task id, a package-minted dispatch number, decision, effect) so a host can tell which in-flight decision an outcome belongs to without searching the prose; the prose is byte-identical, the number is minted only for a decision the in-flight latch admits, and a caller that passes no metadata gets no key. |
+| `scripts/run-plan-review-decide-verify-test.mjs` | `decidePlanReview`'s post-decide honesty ([2315]/[2316], engine RB-471 family): a 2xx from the decide endpoint is **not** a terminal — the wire re-pulls the task status and words the outcome by the real shape (still-locked / legal new gate / genuinely left park / unverified), never claiming success it hasn't earned; when the engine answers that the session's stored resume context cannot be read, the outcome names the unreadable row and says the decision was not applied. Driven against a real fake-engine HTTP server through the shipped dist. The outcome queue item also carries a machine-readable `_sema_planReviewOutcome` (task id, a package-minted dispatch number, decision, effect) so a host can tell which in-flight decision an outcome belongs to without searching the prose; the prose is byte-identical, the number is minted only for a decision the in-flight latch admits, and a caller that passes no metadata gets no key. A plan-review card answered with free text and no option picked (or on the reject option with a note) is read as a rejection carrying that text as its reason, on both the first card and a reopened card; the reason is narrowed by the same step as an approval-denial reason (trimmed, blank dropped, capped at 4096 characters) and a note on the card wins over any reason the caller supplied. An approval with a note stays an approval and sends no note; an empty answer, a contradictory multi-select, several questions or an option outside the shown card stay dismissed; an answer without a note produces a request body byte-identical to before. |
 | `scripts/run-shell-gate-durable-allow-test.mjs` | #110: the durable approval leg for **shell** gates. The tool_end HOLD/REJECT predicate must cover Bash the same way park detection already does (otherwise the park poison frame `Operation aborted` hits the transcript, `endedCalls` swallows the real replayed result, and the user who pressed Yes watches a command that really ran be reported as aborted); a replayed, already-decided park must resume reading the stream instead of being reported as a failed turn; `lastEventId` must track numeric `seq` too. Mutation-proven: each of the three fixes reverted turns the gate red |
-| `scripts/run-hitl-gate-honesty-test.mjs` | [2393] the four HITL disciplines that a passing type-check cannot see. (1) The park predicate and the `tool_end` predicate must cover the **same** set — the park side admits a first-class `kind:'tool_approval'` gate for *any* tool name, and a `tool_end` frame carries no `kind`, so the frame-level judge falls back to the engine's exact abort marker; otherwise the poison frame hits the transcript and `markEnded` swallows the real replayed result (the #110 disease, reopened on kind-only gates). (2) The already-decided identity criterion is **one-shot**: its two inputs are monotonic, so without consumption one successful decide makes every later park failure — including a real `approvals.list` outage — read as "already resolved" until the 24-hop budget runs out and reports a cause that has nothing to do with what happened. (3) A `plan_review` card dismissed without an answer must be re-presentable: the idempotent re-arm short-circuit re-publishes the still-armed card, and a stale armed id (responder gone) re-arms from scratch rather than presenting a card nobody can answer. (4) `HitlSafetyError` is a safety signal — the `remember` fallback arm must re-raise it instead of auto-retrying the decide, while a plain unknown-key 400 still falls back. (5) The polling leg reschedules after an escaping throw and flips `mode()` to `idle` once it consistently fails, so the honesty surface stops reporting a dead feed as live. (6) The live-frame leg carries the fact behind "you are being asked because the auto-mode classifier could not run" all the way to the card port. Transit narrows on SHAPE only — a non-empty cause string is taken verbatim, an open set, because the word table's owner is the engine and re-checking a closed table at the package boundary would drop a legal value the day a new cause word appears, which is exactly the information worth keeping. A malformed carrier degrades to absence rather than half-minting, and absence stays absence: it covers "the classifier answered", "this ask never qualified" and "this deployment has no classifier" at once, so nothing may render it as reassurance. The guard also pins the division of labour that makes the open set safe — the same word that transits is judged again by the public display reader, which narrows to the availability axis, so a word the engine says it never stamps on this fact renders no sentence while still being visible on the card for triage. A later section pins the split this release introduced on the deny close-out frame. Until now every denied tool call was stamped with the same sentence — the one that says *the user* does not want to proceed — including the calls denied automatically on a lane that has no approval surface at all, where nobody was ever asked. The guard drives all three shapes (a person pressed No, a rule settled it, nobody said which) through both close-out arms and the durable park leg, and pins that the third shape is byte-identical to the previous release: an attribution nobody supplied is not evidence for either answer. The rule-settled shape carries the shell's own reason on a second line when there is one and stands alone when there is not, because a blank line where a reason should be reads worse than no line at all. The attribution is read from own data properties only, so neither a polluted prototype nor a getter can make an automatic denial claim a person made it — and the getter case is pinned to never run at all. The transcript classification word is minted only on the two paths where the upstream transcript format really carries one; the three classifier words and the two abort words are left absent, with the abort words pinned against the strings this package actually normalises interruptions to, which are different strings. A final section pins the decide-operation observer: one `start` in the same tick as the first request and exactly one `end` after the last attempt has settled, across success, retried timeouts, exhausted transient failures, semantic refusal, binding mismatch and both kinds of caller abort, with nothing between retries; an observer that throws or rejects — even when logging that fault fails — never changes what is sent or returned, and the stream-level dependency reaches both durable park legs. A package-internal re-delivery of the same decision (plain approve after an older server rejects the session-scope flag, or a re-send without the attribution key) is reported as one operation with a single start and end. An operation handle only groups sends for the same session and bound call — a send for another gate through the same handle is its own operation — and closing a handle while a send is still in flight defers the end until that send settles. When a person picks "allow for this session" on a parked card and the grant is known not to have been stored — the server answers so, which newer servers do for the gates they can recognise from the parked row as needing a person each time, or the server refuses the session-wide grant with one of the refusal codes that are fixed by the row or the deployment and the package falls back to a plain approval — the parked path now says so with the same line the live path uses, and the receipt carries the server's bit for hosts that call that path directly; an answer without the bit, or any other failure — including a conflict that an internal retry can hit after an earlier send already stored the grant — is treated as unknown and says nothing, a plain approval that never asked for the grant says nothing, and a host logger that throws after a successful decision, on either card path or in the bridge's retry step, can no longer turn it into a second send or a failure. From 0.83.0 it also pins approval cards whose arguments are not the tool's real input: a live frame whose arguments were omitted over the size cap or never sent (the card holds at most a path recovered from the question), and a parked approval whose stored input is missing or replaced by the upstream size marker. On every such card an edited approval sends nothing at all — no respond, no decide, no session-grant attempt — ends as edit-refused, raises the same notice once and flags the card so hosts do not offer editing; both production entry points are driven end to end, and the parked path neither mistakes it for an already-decided gate nor reconnects. The explanatory line is pinned word for word in its four forms: a recovered path is the only thing it claims to have recovered, an unrecoverable one says nothing was recovered, neither claims a reconstructed diff, and the parked forms each say which of the two gaps it is, with near-miss shapes of the size marker still rendered as ordinary input. Real input on the stream, the frame or the record keeps edits flowing, and plain approvals and denials are untouched. Since 0.83.2 the live frame's `mandate` word reaches the card only when it is one of the six known words, read back identically by `readApprovalMandate`; it never adds the separate mandated key to the card, but a known word on its own makes `approvalIsMandated` answer true, because the engine only sends the word as the whole reason for that bit. The frame's `origin` is read from its own keys only, so a value inherited through the prototype chain never reaches the card. The already-resolved verdict a decision leg uses (`isAlreadyResolvedFailure`: the closed failure code first; codes that say the gate is still standing or failed to start never fall through to the wording check; the moved-gate codes by code; then the four-word wording fallback) is exported from the root entry as the same function the package calls, and a malformed or hostile input answers `false` without throwing, since a `true` would let a caller silently re-attach. The same section compares it, input by input, with the copy a client kept of the same combination, and requires every disagreement to come from one of the four extra checks that copy adds. It also replays the documented replacement order (the closed failure code first, answered by this function; then the copy's extra checks; then this function) and requires it to reproduce the copy's answers on every input, while moving the extra checks ahead of the closed code must disagree somewhere. A refusal that says the stored form of a parked gate cannot be continued while the gate is still pending is classified by its code and its `retriable` bit as a gate that is still standing, so failure text that happens to contain words from the wording fallback no longer makes the leg re-attach; both decision legs carry the bit on their failure outcome. |
+| `scripts/run-hitl-gate-honesty-test.mjs` | [2393] the four HITL disciplines that a passing type-check cannot see. (1) The park predicate and the `tool_end` predicate must cover the **same** set — the park side admits a first-class `kind:'tool_approval'` gate for *any* tool name, and a `tool_end` frame carries no `kind`, so the frame-level judge falls back to the engine's exact abort marker; otherwise the poison frame hits the transcript and `markEnded` swallows the real replayed result (the #110 disease, reopened on kind-only gates). (2) The already-decided identity criterion is **one-shot**: its two inputs are monotonic, so without consumption one successful decide makes every later park failure — including a real `approvals.list` outage — read as "already resolved" until the 24-hop budget runs out and reports a cause that has nothing to do with what happened. (3) A `plan_review` card dismissed without an answer must be re-presentable: the idempotent re-arm short-circuit re-publishes the still-armed card, and a stale armed id (responder gone) re-arms from scratch rather than presenting a card nobody can answer. (4) `HitlSafetyError` is a safety signal — the `remember` fallback arm must re-raise it instead of auto-retrying the decide, while a plain unknown-key 400 still falls back. (5) The polling leg reschedules after an escaping throw and flips `mode()` to `idle` once it consistently fails, so the honesty surface stops reporting a dead feed as live. (6) The live-frame leg carries the fact behind "you are being asked because the auto-mode classifier could not run" all the way to the card port. Transit narrows on SHAPE only — a non-empty cause string is taken verbatim, an open set, because the word table's owner is the engine and re-checking a closed table at the package boundary would drop a legal value the day a new cause word appears, which is exactly the information worth keeping. A malformed carrier degrades to absence rather than half-minting, and absence stays absence: it covers "the classifier answered", "this ask never qualified" and "this deployment has no classifier" at once, so nothing may render it as reassurance. The guard also pins the division of labour that makes the open set safe — the same word that transits is judged again by the public display reader, which narrows to the availability axis, so a word the engine says it never stamps on this fact renders no sentence while still being visible on the card for triage. A later section pins the split this release introduced on the deny close-out frame. Until now every denied tool call was stamped with the same sentence — the one that says *the user* does not want to proceed — including the calls denied automatically on a lane that has no approval surface at all, where nobody was ever asked. The guard drives all three shapes (a person pressed No, a rule settled it, nobody said which) through both close-out arms and the durable park leg, and pins that the third shape is byte-identical to the previous release: an attribution nobody supplied is not evidence for either answer. The rule-settled shape carries the shell's own reason on a second line when there is one and stands alone when there is not, because a blank line where a reason should be reads worse than no line at all. The attribution is read from own data properties only, so neither a polluted prototype nor a getter can make an automatic denial claim a person made it — and the getter case is pinned to never run at all. The transcript classification word is minted only on the two paths where the upstream transcript format really carries one; the three classifier words and the two abort words are left absent, with the abort words pinned against the strings this package actually normalises interruptions to, which are different strings. A final section pins the decide-operation observer: one `start` in the same tick as the first request and exactly one `end` after the last attempt has settled, across success, retried timeouts, exhausted transient failures, semantic refusal, binding mismatch and both kinds of caller abort, with nothing between retries; an observer that throws or rejects — even when logging that fault fails — never changes what is sent or returned, and the stream-level dependency reaches both durable park legs. A package-internal re-delivery of the same decision (plain approve after an older server rejects the session-scope flag, or a re-send without the attribution key) is reported as one operation with a single start and end. An operation handle only groups sends for the same session and bound call — a send for another gate through the same handle is its own operation — and closing a handle while a send is still in flight defers the end until that send settles. When a person picks "allow for this session" on a parked card and the grant is known not to have been stored — the server answers so, which newer servers do for the gates they can recognise from the parked row as needing a person each time, or the server refuses the session-wide grant with one of the refusal codes that are fixed by the row or the deployment and the package falls back to a plain approval — the parked path now says so with the same line the live path uses, and the receipt carries the server's bit for hosts that call that path directly; an answer without the bit, or any other failure — including a conflict that an internal retry can hit after an earlier send already stored the grant — is treated as unknown and says nothing, a plain approval that never asked for the grant says nothing, and a host logger that throws after a successful decision, on either card path or in the bridge's retry step, can no longer turn it into a second send or a failure. From 0.83.0 it also pins approval cards whose arguments are not the tool's real input: a live frame whose arguments were omitted over the size cap or never sent (the card holds at most a path recovered from the question), and a parked approval whose stored input is missing or replaced by the upstream size marker. On every such card an edited approval sends nothing at all — no respond, no decide, no session-grant attempt — ends as edit-refused, raises the same notice once and flags the card so hosts do not offer editing; both production entry points are driven end to end, and the parked path neither mistakes it for an already-decided gate nor reconnects. The explanatory line is pinned word for word in its four forms: a recovered path is the only thing it claims to have recovered, an unrecoverable one says nothing was recovered, neither claims a reconstructed diff, and the parked forms each say which of the two gaps it is, with near-miss shapes of the size marker still rendered as ordinary input. Real input on the stream, the frame or the record keeps edits flowing, and plain approvals and denials are untouched. Since 0.83.2 the live frame's `mandate` word reaches the card only when it is one of the six known words, read back identically by `readApprovalMandate`; it never adds the separate mandated key to the card, but a known word on its own makes `approvalIsMandated` answer true, because the engine only sends the word as the whole reason for that bit. The frame's `origin` is read from its own keys only, so a value inherited through the prototype chain never reaches the card. The already-resolved verdict a decision leg uses (`isAlreadyResolvedFailure`: the closed failure code first; codes that say the gate is still standing or failed to start never fall through to the wording check; the moved-gate codes by code; then the four-word wording fallback) is exported from the root entry as the same function the package calls, and a malformed or hostile input answers `false` without throwing, since a `true` would let a caller silently re-attach. The same section compares it, input by input, with the copy a client kept of the same combination, and requires every disagreement to come from one of the four extra checks that copy adds. It also replays the documented replacement order (the closed failure code first, answered by this function; then the copy's extra checks; then this function) and requires it to reproduce the copy's answers on every input, while moving the extra checks ahead of the closed code must disagree somewhere. A refusal that says the stored form of a parked gate cannot be continued while the gate is still pending is classified by its code and its `retriable` bit as a gate that is still standing, so failure text that happens to contain words from the wording fallback no longer makes the leg re-attach; both decision legs carry the bit on their failure outcome. The judge of whether a remember request was refused by the server specifically is a root export — it answers the matching code for the five dedicated codes and nothing otherwise, never throws, and agrees with the durable approval leg code for code. |
 | `scripts/run-park-hop-progress-test.mjs` | L-80: the park re-attach loop budgets **stalled** rounds, not parks. A turn where the model keeps hitting gates and every one of them is really decided (a card was answered, the engine really moved on) must never be cut off by the hop budget — the budget counts consecutive rounds that produced no progress, and "the engine revived and immediately parked again on the same coordinates" is not progress. The three non-progress arms (already-resolved, decide-transport-exhausted, and a re-scan that was adopted but led nowhere) share one same-cause limit instead of one arm having a limit and the others having none, and every non-progress re-attach is announced once through the host callback rather than only to the debug log. When the limit is spent the resolver reads the approval queue once more and puts whatever is decidable in front of the user before it gives up; only when there is genuinely nothing to show does it fail soft, and the terminal message then carries the real cause and a real way out instead of a sentence about a budget. On the self-heal side, a reopen verdict that reports `decidedWithoutCard` — the chain settled the gate by rule, so there was no card to present — is progress, not a reopen failure, and the user is not told their message was NOT sent. Negative control: a genuinely empty queue with a run that never moves still fails soft |
 | `scripts/run-notif-fleet-honesty-test.mjs` | [2393] the five notification/fleet disciplines a green type-check cannot see, each proven by reverting the fix. (1) The workflow-side dedup `return` keeps a count and a trace — without it "suppressed by design" and "a real completion swallowed because the runId minting changed" are the same observation. (2) `seq` normalisation has exactly one mint point, so a 0-based or fractional wire `seq` cannot make the watcher lane and the frame lane key the same completion differently (which would feed the model twice). (3) The TTL sweep defers to a probe arm that is still inside its own deadline — an entry recorded as "abandoned" must not be delivered a moment later — while an arm that has outlived its deadline never blocks the sweep, so the headless exit gate keeps its liveness. (4) The reset hook really clears every ledger it claims to (the sticky `prompt` ledger leaked across cases). (5) The fleet ledger counts all three drop paths (malformed / unknown frame type / isolation drop), and the panel projection's settled recycling is anchored on the settle instant and skips still-present rows, so the dedup token is never carried off with the entry (which would re-emit `end`) |
 | `scripts/run-public-surface-test.mjs` | The outward promises: the npm export surface baseline (an **exact set**, both directions — a new export that never entered the baseline is one nobody watched leave, and deleting it later would not be red), the peer floor witness, and this README's claims. The floor witnesses are a list since 0.90.0 (the installed SDK plus the 15.x and 16.0.0 probe trees); every listed SDK is checked for the value-level symbols and the cost field, and a missing 16.0.0 witness is reported as a skipped section rather than as green. |
@@ -373,11 +373,11 @@ guard still cross-checks the table by name).
 | `scripts/run-package-hygiene-test.mjs` | Everything `package.json` `files` ships — dist JS/typings and the Markdown docs — is screened line-by-line against a deny-list of strings that must never reach a public tarball (internal hostnames, codenames, person names, collaboration-process words, other repos' ledger ids and repo names; opaque ticket ids `CC-nnn` and post numbers `[nnnn]` are allowed as traceability references). Since 0.77.2 the build strips comments (`removeComments`; enforced by `run-dist-comments-test.mjs`), so what this gate screens in dist is code, string literals and type-level text. Markdown docs are enforced forward-only (CHANGELOG from 0.77.2, the integration doc from §81) because published sections are frozen. From 0.88.0 every shipped file is also scanned in full, frozen sections included, for merge-conflict markers at the start of a line (`<<<<<<< `, a bare `=======`, `>>>>>>> `, and the seven-bar base marker of three-way conflicts). From 0.88.0 a short local list of release-process phrasings (integration-branch and review-round wording) is screened as well, forward-only from the 0.88.0 sections.
 | `scripts/run-integration-doc-freshness-test.mjs` | The **integration contract** (`docs/INTEGRATION-CLIENTS.md`) and the **changelog** (`CHANGELOG.md`) checked against the code, because a document with no guard rots — this one had a whole nest of drift found on it within a day of being written. Five directions, each a claim a machine can actually evaluate. (1) *Counting discipline*: the version-anchor row for the guard count may no longer carry a hand-copied number at all — it changes every time a guard is added, and writing it down is planting a timer; the export counts that are still hand-copied (the surface total, the test-hook count, the sentence describing the surface's internal composition, the sum of the sixteen domain rows, and the three sub-counts) are each compared against a value **derived** from `public-export-baseline.json`, which is the drift a human reviewer caught last time. (2) *Coordinates alive*: every `src/` `scripts/` `docs/` path the doc quotes must be on disk **and tracked by git** — on disk is not in the repo, and a doc that points readers at a file living only in its author's working tree sends every clone to nothing. A file landing in the same commit takes a named carve-out that **stops applying** the moment the file is really tracked (it can no longer let anything through, and the guard prints a line asking for it to be deleted) — deliberately not a red, since turning red on the very commit that lands the file would just manufacture a break that only a follow-up commit could clear. (3) *Arm tables*: the `hitl_out_of_slice` row and the `not_in_slice` fenced list must equal, name for name and in **both** directions, the case labels that really fall into those two buckets — read through the **TypeScript AST**, since which bucket an arm lands in is decided by the argument to `nothing(...)` and by nothing a comment says. The extractor is anchored to the one production projector: exactly one function named `eventToSdkMessage`, exactly one `switch (ev.type)` inside it, and no repeated case label — anything else is a broken anchor rather than a verdict, because a second same-shaped switch elsewhere in the file would otherwise overwrite the real one's conclusions and leave the doc agreeing with a switch nobody runs. The list is delimited by a machine-readable fence rather than by section headings, because the same section also names the terminal arms as a counter-example and prose boundaries cannot tell a member from a foil. (4) *Released sections are frozen*: an **append-only ledger** carries every version ever published — its number, the commit it was published from, and the sha256 of its section — and each one is checked, not just the current release, since pinning only the latest would set every earlier version free the moment the next one ships. The ledger cannot vouch for itself either: each recorded hash is **re-derived from that release commit** through git, so editing an old section and its constant together no longer passes — and the commit the row names is in turn checked against the `gitHead` npm recorded at publish time, which is the one value this repository cannot rewrite, so pointing an old version at a freshly written commit does not pass either. The *set* of versions that must be frozen comes from the registry too, so deleting an old row together with its section — which would otherwise remove that version from every set the guard looks at — is red rather than invisible. A failed registry call is classified rather than swallowed, and the classification consults the registry's own status code *before* it considers connection-level symptoms, so an auth refusal whose body happens to mention the network is still red rather than a skip. The version set is compared as full SemVer including prereleases — matching only `x.y.z` would silently drop a published `0.30.0-beta.1` and reopen the very hole this direction closes — and section headings are matched on a whole-version boundary so a stable release cannot bind itself to the release-candidate section sitting above it. Publishing itself is a two-phase protocol rather than a paradox: before a release, exactly one row may be marked pending and must name the current `package.json` version, exempt from the checks whose inputs do not exist yet; once the registry has that version the row must be promoted, so the temporary state cannot survive its own release. And because the pending exemption rests entirely on "this version is not out yet," it is refused outright when the registry cannot be reached to confirm that — an unverifiable premise is not a licence. Three reverse directions close the rest: a section claiming to be released but absent from the ledger, a ledger entry whose section has vanished, and a `package.json` version that was never frozen. Publishing appends a row; it never rewrites one. (6) *Sentinels*: the readers §5a hands hosts for "is this port installed" are checked against what the source actually declares it returns — `hasXxx()` is a `boolean`, the card port / HITL surface / wire target return `T \| null`, the `installHost` family returns `T \| undefined`. Testing a `null`-returning reader for `!== undefined` is *always true*, and a self-check that passes whether or not the port is installed is worse than none, because hosts retire their own fallback on the strength of it. Both directions are red: an implementation that changes its sentinel without the doc following, and a doc that names the wrong one. The roster covers the zero-argument readers and their `*For` variants alike — a multi-session host reads the variants, so leaving them off would let exactly the surface desktop depends on drift unwatched — and the §5a table and the §8-B checklist line are each checked against the source, because hosts tick the checklist, and a guard that only watches the prose table misses the line people actually follow. (5) *Packaging*: the README ships with the package and opens by pointing hosts at the integration doc, and the checklist names two more files as required reading before an upgrade — all three must really appear in the `npm pack` manifest, or an npm consumer follows a relative link that npmjs rewrites onto a private repository. Missing tooling never takes the whole verdict down with it: when git, npm or the registry is unreachable those legs print the `SKIPPED-SECTION` marker and the rest still judges, while a release commit the ledger names but git cannot resolve is red rather than skipped. The guard says in its own header what it does **not** do: it judges counts, coordinates, arm sets, released bytes and the packing list — whether a sentence is *right* is still for review and for the hosts to report (7) *Retired names*: every name in the per-version `removed` ledger of `scripts/export-liveness.json` may appear in the live sections of the integration doc only where a retirement note follows the name inside the same clause (or the table row's label cell is itself a retirement label); the scan is by identifier boundary after invisible text (HTML comments, link targets, reference-link labels, tag attributes) has been stripped, so a signature line in a code block, an inline `NAME = 4096`, a hidden note, or a note that belongs to a neighbouring name all count as a bare recommendation and go red. Frozen sections (a numbered section whose heading carries a version already recorded as released) are historical and never rewritten, so a retired name there is allowed only if the live retirement catalog has a row for it: a retirement label, the name, the version it left in (matching the ledger) and what to use instead, next to the sentence stating that frozen sections are historical records and the catalog is authoritative. A section whose version cannot be read, or is not yet released, is judged as live. (9) *Known-limits tables*: every row of every table in `docs/KNOWN-LIMITS.md` must split into as many cells as its header — a pipe character inside inline code still splits a Markdown table cell unless it is escaped, which silently shifts the trigger and upgrade columns. |
 | `scripts/run-type-superset-ledger-test.mjs` | The type/wire **superset ledger** (`docs/type-superset.json`): positions this package adds on top of a CC-shaped contract, each carrying the evidence for what CC's own type surface does or does not have there. Completeness is deliberately uneven and the ledger says so. The `_sema_*` private-key class is checked in **both** directions (a key in the source that never entered the ledger is red, naming key and file; a ledger row whose key left the source is red) — but only for keys written as literals, which is the convention the ledger mandates. A key assembled by string arithmetic is beyond what any static rule can enumerate, so the guard fails closed on every shape it *can* decide (a bare `_sema_` prefix is red wherever it appears, save one pinned guard site) and leaves the rest as a convention violation for review to catch, rather than claiming a completeness it does not have. The two hand-surveyed classes are only checked for coordinate and evidence integrity, never discovered. Both directions read the source through the **TypeScript AST**, not a text scan, and they read two different sets out of it. A *key site* is an identifier, or a string whose whole value is the key — so `'_sema_decision-v2'` is carried whole rather than truncated at the first non-identifier character into some *other* key that happens to be registered. A *mention* is the key appearing inside a longer string, which is prose, not usage. The staleness direction counts key sites only: a comment or a doc sentence left behind after the last real mint site is deleted must not keep the row alive (mutation-proven — with both the comment and the prose string untouched, removing the one real site turns the guard red). And because a prefix can be concatenated or interpolated into a key no static set will ever see, the bare `_sema_` literal is refused outright rather than traced: every occurrence is red except the single inline `startsWith` guard the sanitizer needs, because the set of expressions a bare prefix can travel through on its way to a concatenation is open-ended and enumerating it is always one form behind. Every row's `host` must still resolve, with the key being a real **member of that declaration** rather than a string occurring somewhere in the same file — `governanceForced`/`delegation` each live on two different shapes in one file, and a member commented out is a member deleted, which a text-shaped check happily reads as still present. And the direction worth the most: each machine-form `ccAbsenceEvidence` is re-derived from the row's own `key` — the ledger's recorded string must match that derivation verbatim, since a row quietly witnessing `\bnever_present\b` is green forever while watching nothing (mutation-proven: the same edit passes the unbound form and is caught by the bound one) — and the check runs against the names the installed `@sema-agent/agent-types` `.d.ts` set actually declares, parsed with the TypeScript AST rather than grepped, so a name CC merely mentions in a comment cannot force the row into the manual escape hatch and thereby retire the very witness that was supposed to fire the day CC declares that name for real. That escape hatch is gated by an allowlist living **in the guard**, not the ledger, so claiming it costs a reviewed diff. Missing material never reads as a pass, and the verdict splits by *why* it is missing: no TypeScript parser skips the suite before it starts; a missing `agent-types` still runs and prints the first three directions, then exits **1** when `package.json` declares the mirror but it is not installed — a broken install must not retire the repository's only "the day CC declares this name" alarm, and reporting it as a skip would leave "never evaluated" and "evaluated, no drift" indistinguishable to the runner — and exits 3 only when nothing declares the mirror at all, which is the one case where the direction genuinely does not apply. Either way a run that evaluated no witness is never counted as one that did. When the mirror *is* present its **installed version** is witnessed too (the two declared floors must agree with each other and the installed copy must meet them), since four preflight probes are satisfied by an arbitrarily stale mirror — they prove the extractor speaks, not that it is current. Every direction carries a positive control — known-present CC symbols, a comment-only sample proving the extractor distinguishes declaration from mention, and synthetic corpora fed through the **same** discriminator function the real verdict uses, so a verdict quietly rewritten to return nothing takes its own control down with it |
-| `scripts/run-rules-side-test.mjs` | The persisted-permission-rules lane's shared decision half. From 0.92.0 the governance surface reads one capability bit, the rules lane: every server at or above the support floor that advertises it also serves the revoke routes, so the second bit is no longer consulted and a lane-only body shows the surface. Failure classification is by **disposition, not cause**: the two 404s (route missing vs. dead ticket) never share a bucket, a 503 `rule_import_retry` means *the ticket is still alive* (the opposite handling of a dead one), and a stale-cursor 400 drops the cursor and re-lists from the top exactly once — never resuming a stale keyset, never surfacing a partial governance list, and never paging past the hard cap. The persist-ack reader is **merged into** `readToolApprovalRespondAck`: the three-state verdict (`persisted` / `refused` / `unknown`) is derived only from an ack that passed the package's structural narrowing, and a half-shaped object such as `{rulePersisted: true}` with no `delivery` reads as `unknown` — the pre-merge shell read would have said `persisted`, which is precisely the double-ledger drift this file closes, so that case is pinned in reverse. The local-allow-rule skeleton pins all five narrowings (whole-tool, tool-name match, literal anchor with the escaped-star counter-example, bare interpreter prefix consulted only for Bash, and the canonical dangerous-pattern overlay) **with their refusal strings byte-for-byte** — the cli's 128-assertion suite anchors the same strings, so a one-character edit here changes observable behaviour on three clients — and asserts the parse is a pure function of its input, because the same call backs both "render the option" and "resolve the selected value" `listAllPersistedRules` needs only `list` (`RulesListFacade`; the other four methods are known to the parameter type as optional members): a synthetic consumer compiled against the built declarations passes a two-method object, a list-only literal, a `Pick` slice, the named type, the full facade, and the full or two-method facade written as an inline object literal — including arrow functions with untyped parameters — all with zero diagnostics, while a misspelled method name in such a literal is still reported. From 0.94.0 a `request.query_invalid` answer is split by whether that request carried a cursor: with a cursor it is still a stale cursor, without one the listing's query itself was refused (`query-refused`, for example a project scope that is not an absolute path), the server's sentence is kept and the listing is not retried. |
+| `scripts/run-rules-side-test.mjs` | The persisted-permission-rules lane's shared decision half. From 0.92.0 the governance surface reads one capability bit, the rules lane: every server at or above the support floor that advertises it also serves the revoke routes, so the second bit is no longer consulted and a lane-only body shows the surface. Failure classification is by **disposition, not cause**: the two 404s (route missing vs. dead ticket) never share a bucket, a 503 `rule_import_retry` means *the ticket is still alive* (the opposite handling of a dead one), and a stale-cursor 400 drops the cursor and re-lists from the top exactly once — never resuming a stale keyset, never surfacing a partial governance list, and never paging past the hard cap. The persist-ack reader is **merged into** `readToolApprovalRespondAck`: the three-state verdict (`persisted` / `refused` / `unknown`) is derived only from an ack that passed the package's structural narrowing, and a half-shaped object such as `{rulePersisted: true}` with no `delivery` reads as `unknown` — the pre-merge shell read would have said `persisted`, which is precisely the double-ledger drift this file closes, so that case is pinned in reverse. The local-allow-rule skeleton pins all five narrowings (whole-tool, tool-name match, literal anchor with the escaped-star counter-example, bare interpreter prefix consulted only for Bash, and the canonical dangerous-pattern overlay) **with their refusal strings byte-for-byte** — the cli's 128-assertion suite anchors the same strings, so a one-character edit here changes observable behaviour on three clients — and asserts the parse is a pure function of its input, because the same call backs both "render the option" and "resolve the selected value" `listAllPersistedRules` needs only `list` (`RulesListFacade`; the other four methods are known to the parameter type as optional members): a synthetic consumer compiled against the built declarations passes a two-method object, a list-only literal, a `Pick` slice, the named type, the full facade, and the full or two-method facade written as an inline object literal — including arrow functions with untyped parameters — all with zero diagnostics, while a misspelled method name in such a literal is still reported. From 0.94.0 a `request.query_invalid` answer is split by whether that request carried a cursor: with a cursor it is still a stale cursor, without one the listing's query itself was refused (`query-refused`, for example a project scope that is not an absolute path), the server's sentence is kept and the listing is not retried. Since 0.95.0 a caller that cannot say whether its request carried a cursor (no second argument, a non-boolean, or a context whose read throws) gets the generic error with the server's sentence instead of a stale-cursor claim: re-listing from the top is useless for a request that carried no cursor, so naming that cause without evidence is withheld. |
 | `scripts/run-park-decision-layer-test.mjs` | The decision layer behind the "stuck behind a card" family, shared by every client. A pending row that is **not in the queue** is three states, not one: a bounded, interruptible re-probe loop distinguishes *a decidable row*, *not born yet* (no positive evidence that anything settled — an empty queue proves nothing) and *settled elsewhere*, always probes at least once so a zero budget keeps the pre-fix semantics verbatim, cuts a hung read face off at the window rather than only noticing afterwards, and reports the honest failure when the window is spent instead of inventing a decision. The decision-note reader is likewise three-state: an explicit `noteRecorded: false` outranks an echoed note body, absence renders **no line at all**, and untrusted note text is flattened and bounded before it ever reaches a renderer. Row routing anchors on the deciding quantity — a row carrying `gateKind: "human"` with `toolName: "Write"` is a tool gate, because `human` is the engine's *generic* "someone must decide", not a synonym for a question — and the queue scan refuses to surface a row it cannot positively prove belongs to this session. A chain that fails after the row vanished is split by whether a card was ever presented: decided-elsewhere, or not-its-turn-yet. A row-level single-flight makes "at most one card per pending item" structural rather than incidental. The resume three-way card pins the option **order** (the zero-effect choice sits at index 0, because the frame carries no default-focus field and a stray Enter must not attach or cancel), renders only options the wired verbs can honour, collapses every ambiguous answer to zero action, omits the liveness line entirely when the engine gave no evidence, and — when there is no card lane at all — prints three real routes and exits on a dedicated code rather than reporting success |
-| `scripts/run-selfheal-reopen-test.mjs` | The 409 active-run self-heal decision chain: `governanceForced` narrows on strict `true` only; triage prefers the wire's `pendingGate.kind` and falls back to the status table (an off-table kind is never guessed into a card arm — hands-off plus the honest wording); a first-sight card makes zero closed/reopened claims and a host presentation receipt of `presented: false` demotes the outcome to reopen-failed; park-row ownership is a fail-closed positive proof (own-run ledger or session id — unprovable is not owned); the three gate-identity key literals live in exactly one mint (`hitl/gateIdentity.ts`, AST string-token scan); the armed-gate presentation ledger is per-session; and the `plan_review` reopen arm shares the arm arm's card body, three-state verdict and delivery pipe, consuming the presentation history once a decision is delivered. The same chain also carries the `running` three-way card: both plan-family gate kinds route to the plan arm and all four ask-family kinds to the ask arm (an off-table kind still never gets guessed into either); the card is offered only for verbs that can actually be honoured and a missing presenter means zero action rather than a silent cancel; a steer is sent **exactly once** with its three delivery outcomes worded apart (a `queued` receipt is the wire correcting the triage input, so the named park word decides which card gets reopened, and an unrecognised park word drives neither arm), and a steer failure is split into *provably not delivered* (4xx) and *delivery unknown*, because telling a user to resend a non-idempotent instruction that may already have landed is how duplicates get made. After a user-chosen cancel, "the session is free" is asserted only from a whitelist of terminal states — park states hold the claim, an unrecognised state word is not a release, a failed read is *unknown* rather than a release, and only a 404 counts as one — and the honest timeout line quotes how long it really waited. The two "card could not be reopened" rows can carry a host-declared way to keep the conversation, which says the card comes back on resume only if it is still waiting: it is placed before the route that abandons it, never offered for an injected submission, while a decision is still on its way, or once the pending approval has been proven gone (the outcome then carries a flag saying so; the proof only counts before the cleanup card is shown, so a fallback after the card carries no flag unless a fresh read finds the run finished, and a recheck that finds the approval back clears it), the host function is not even called in those cases, and it is treated as unavailable when it throws or returns an empty value; a host can also switch off the engine decide route on the interactive rows while the cancel route stays, and with neither given all four rows are pinned byte-for-byte to the text the previous release produced. When the plan reopen refuses because the user closed that card (`dismissedByUser`), the outcome carries the flag and one package-owned line says so — you closed it, a message brings it back — in the typed, the injected and the queued-steer forms, with no way-to-keep and no engine decide route; the reopen port receives `{ trigger }` mapped from the submission origin (typed ⇒ `user`, injected ⇒ `automatic`, absent ⇒ the old single-argument call). When the reopen refuses because no view is subscribed to question frames for that session (`_sema_noPresentationSurface`), the plan and ask outcomes carry `noPresentationSurface`, and the five lines that talk about the card not coming back — both arms' typed line, the queued-steer typed and injected lines, and the injected not-delivered line — add one package-owned clause saying the session has no view that can show decision cards, byte-identical to the plain refusal otherwise; loose values and the other refusals (answer in flight, user-closed card, plain refusal) carry neither the flag nor the clause. For a host that reopens under its own session key with a presentation receipt, the suite drives the whole sequence end to end: subscribing for that key, reopening, and registering the receipt through the per-session call `registerArmedGateFromQuestionIdFor` gives `reopened: true` with the right first-sight reading, a second reopen reads as seen before, a decisive answer (with no extra bookkeeping from the host) makes the next reopen read as first sight again, and a dismissal does not; registering through the default-session call instead leaves the receipt window to run out with `{ reopened: false }`. Every bit the package reads off a host's reopen verdict is read once, inside the guard: a bit whose getter throws (or a proxy that throws) is treated as absent, so the self-heal still settles — with exactly the outcome it would give if that bit were missing — and the rows never throw, including for an outcome the host built itself; a plain verdict lands in the queued-steer outcome byte for byte, key order and any extra keys the host added included, a key whose getter throws is the only one left out, and an own key literally named `__proto__` is copied as a key rather than changing the copy's prototype; symbol keys are copied too and a verdict with a null prototype stays one (compared key by key, value by value and by prototype, not through JSON); and an outcome whose `reopened` property itself cannot be read is phrased as if it had none. The two reopen-failed outcomes carry `errorCode` only when the reopen gave up because a request failed — the outer wire code of that request (a folded policy code is read through `blockedBy`), taken from a reopen port that threw, from the host verdict's `_sema_errorCode`, or from the stale-park arm's own three requests; a later request that changed what the reopen gave up on (the poll finally answered, the row appeared), a failure with no wire code, a reopen port that failed after the caller had already aborted, and the three judgment refusals (answer in flight, user-closed card, no view) all leave it absent, while a read that only confirms (the row is still pending after a failed decision) keeps it; only a request that failed counts, never a field that cannot be read on a record that did come back; and every row is byte-identical with or without it. The error objects the self-heal leg reads (thrown by the run lookup, the steer and the cancel) are read inside the guard too: a `status` or `message` whose getter throws, or a proxy that throws on any read, is treated as absent, so the self-heal still settles instead of rejecting, and the at-most-once failure class reads such an error as delivery unknown; the same goes for the run record a successful lookup returns — when its `status` cannot be read, the self-heal treats it as a record without a status rather than rejecting — and for a steer receipt, whose two readers answer `null` for a field they cannot read. When the delivery of a steer is missing, unreadable or an unknown word, the injected-submission line says the message was accepted but its delivery was not reported and to watch that reply before deciding whether anything needs sending again, instead of the handed-over wording, and it never says nothing is needed, with or without a follow declaration. |
+| `scripts/run-selfheal-reopen-test.mjs` | The 409 active-run self-heal decision chain: `governanceForced` narrows on strict `true` only; triage prefers the wire's `pendingGate.kind` and falls back to the status table (an off-table kind is never guessed into a card arm — hands-off plus the honest wording); a first-sight card makes zero closed/reopened claims and a host presentation receipt of `presented: false` demotes the outcome to reopen-failed; park-row ownership is a fail-closed positive proof (own-run ledger or session id — unprovable is not owned); the three gate-identity key literals live in exactly one mint (`hitl/gateIdentity.ts`, AST string-token scan); the armed-gate presentation ledger is per-session; and the `plan_review` reopen arm shares the arm arm's card body, three-state verdict and delivery pipe, consuming the presentation history once a decision is delivered. The same chain also carries the `running` three-way card: both plan-family gate kinds route to the plan arm and all four ask-family kinds to the ask arm (an off-table kind still never gets guessed into either); the card is offered only for verbs that can actually be honoured and a missing presenter means zero action rather than a silent cancel; a steer is sent **exactly once** with its three delivery outcomes worded apart (a `queued` receipt is the wire correcting the triage input, so the named park word decides which card gets reopened, and an unrecognised park word drives neither arm), and a steer failure is split into *provably not delivered* (4xx) and *delivery unknown*, because telling a user to resend a non-idempotent instruction that may already have landed is how duplicates get made. After a user-chosen cancel, "the session is free" is asserted only from a whitelist of terminal states — park states hold the claim, an unrecognised state word is not a release, a failed read is *unknown* rather than a release, and only a 404 counts as one — and the honest timeout line quotes how long it really waited. The two "card could not be reopened" rows can carry a host-declared way to keep the conversation, which says the card comes back on resume only if it is still waiting: it is placed before the route that abandons it, never offered for an injected submission, while a decision is still on its way, or once the pending approval has been proven gone (the outcome then carries a flag saying so; the proof only counts before the cleanup card is shown, so a fallback after the card carries no flag unless a fresh read finds the run finished, and a recheck that finds the approval back clears it), the host function is not even called in those cases, and it is treated as unavailable when it throws or returns an empty value; a host can also switch off the engine decide route on the interactive rows while the cancel route stays, and with neither given all four rows are pinned byte-for-byte to the text the previous release produced. When the plan reopen refuses because the user closed that card (`dismissedByUser`), the outcome carries the flag and one package-owned line says so — you closed it, a message brings it back — in the typed, the injected and the queued-steer forms, with no way-to-keep and no engine decide route; the reopen port receives `{ trigger }` mapped from the submission origin (typed ⇒ `user`, injected ⇒ `automatic`, absent ⇒ the old single-argument call). When the reopen refuses because no view is subscribed to question frames for that session (`_sema_noPresentationSurface`), the plan and ask outcomes carry `noPresentationSurface`, and the five lines that talk about the card not coming back — both arms' typed line, the queued-steer typed and injected lines, and the injected not-delivered line — add one package-owned clause saying the session has no view that can show decision cards, byte-identical to the plain refusal otherwise; loose values and the other refusals (answer in flight, user-closed card, plain refusal) carry neither the flag nor the clause. For a host that reopens under its own session key with a presentation receipt, the suite drives the whole sequence end to end: subscribing for that key, reopening, and registering the receipt through the per-session call `registerArmedGateFromQuestionIdFor` gives `reopened: true` with the right first-sight reading, a second reopen reads as seen before, a decisive answer (with no extra bookkeeping from the host) makes the next reopen read as first sight again, and a dismissal does not; registering through the default-session call instead leaves the receipt window to run out with `{ reopened: false }`. Every bit the package reads off a host's reopen verdict is read once, inside the guard: a bit whose getter throws (or a proxy that throws) is treated as absent, so the self-heal still settles — with exactly the outcome it would give if that bit were missing — and the rows never throw, including for an outcome the host built itself; a plain verdict lands in the queued-steer outcome byte for byte, key order and any extra keys the host added included, a key whose getter throws is the only one left out, and an own key literally named `__proto__` is copied as a key rather than changing the copy's prototype; symbol keys are copied too and a verdict with a null prototype stays one (compared key by key, value by value and by prototype, not through JSON); and an outcome whose `reopened` property itself cannot be read is phrased as if it had none. The two reopen-failed outcomes carry `errorCode` only when the reopen gave up because a request failed — the outer wire code of that request (a folded policy code is read through `blockedBy`), taken from a reopen port that threw, from the host verdict's `_sema_errorCode`, or from the stale-park arm's own three requests; a later request that changed what the reopen gave up on (the poll finally answered, the row appeared), a failure with no wire code, a reopen port that failed after the caller had already aborted, and the three judgment refusals (answer in flight, user-closed card, no view) all leave it absent, while a read that only confirms (the row is still pending after a failed decision) keeps it; only a request that failed counts, never a field that cannot be read on a record that did come back; and every row is byte-identical with or without it. The error objects the self-heal leg reads (thrown by the run lookup, the steer and the cancel) are read inside the guard too: a `status` or `message` whose getter throws, or a proxy that throws on any read, is treated as absent, so the self-heal still settles instead of rejecting, and the at-most-once failure class reads such an error as delivery unknown; the same goes for the run record a successful lookup returns — when its `status` cannot be read, the self-heal treats it as a record without a status rather than rejecting — and for a steer receipt, whose two readers answer `null` for a field they cannot read. When the delivery of a steer is missing, unreadable or an unknown word, the injected-submission line says the message was accepted but its delivery was not reported and to watch that reply before deciding whether anything needs sending again, instead of the handed-over wording, and it never says nothing is needed, with or without a follow declaration. The `running` three-way card also gives a fourth way out, *attach*, under the same single rule as the resume card (the run is still running and can be taken over): one shared predicate decides for both cards, only a strict `true` supply flag counts, and a host that does not state the flag gets the request byte-for-byte as before (no `canAttach` key at all, which is not the same as `false`). Choosing it hands the decision to the host without sending any write request (no steer, no cancel), the refused message is reported as not sent, and an `attach` answer on a card that never offered it is zero action. The resume card's two cancel closing rows, confirmed and unconfirmed, are mutually exclusive and come from one private mint. |
 | `scripts/run-terminal-identity-copy-test.mjs` | Terminal-state **identity**, in both lanes where a stop gets a name. A run stopped by this deployment's own governance knobs — the open-set `limits.*` family, `output.invalid`, and the `blocked` contract terminal a ReportBlocked agent produces — is not a provider failure, and labelling it `API Error:` sends the reader to check the network, the key and the quota when the handle is the `--max-turns` they passed themselves. Those terminals now render a neutral row; the reverse direction is guarded just as hard, because asserting "this is *not* an API error" on a code the package does not recognise is the same misfiling pointed the other way — a real `gateway HTTP 502`, a `conflict.session_active_run` and any unknown code all keep the `API Error:` prefix, and the row keeps its `_sema_api_error_message` class flag (spelled `isApiErrorMessage` before 0.83.0) so brief-mode visibility filtering does not silently drop it. The second half is who the rejected submission belonged to: the self-heal copy told every caller "Your message was NOT sent … send it again", which is three separate untruths for a system injection (a plan-review outcome, a cron wake-up, a task notification) — not the user's message, and not re-sendable, since a host queue marks those non-editable and non-recallable. The injected form says so instead, and the one sentence that promises re-delivery is pinned to the single disposition that earns it: `selfHealSubmissionDisposition` is the same function the host consults before putting the item back on its queue, so the promise and the behaviour cannot drift apart, and the arms where no card could be surfaced state plainly that nothing was delivered and nothing will retry. Since 0.72.6 the same gate pins the **follow intent** after a steer (): a message handed to a live run only pays off if someone tails that run's own event stream, so `steerFollowIntent` decides from the delivery word whether to tail now, after the pending decision, or only after a wake — and the "watch that run" sentence ("watch that reply" on the rows about follow-up messages sema sent on its own) turns into a factual "sema is following that run" ("… that reply") **only** when the host declares it attached that tail, so a shell that did not wire it can never claim it did. Since 0.83.0 the rows about follow-up messages sema sent on its own carry no engine-internal words and say what happened per delivery shape, promising a resend or "nothing for you to do" only where the code guarantees it |
-| `scripts/run-additive-key-passthrough-test.mjs` | The one disease shape behind two legs: a **closed whitelist / flattening arm** dropping a fact that is already on the wire, while both sides of the seam look correct. (1) The `task_progress` projection carries a registered **key ledger** — a frame populated with every key the service really projects is pushed through the shipped `eventToSdkMessage`, and the set of wire keys that survive must equal the registered pass-through list **name for name in both directions**, so quietly forwarding one more key is as red as quietly dropping one. `model` (the child run's model id, minted by core as `prepared.model.id` and projected by the server since 7.52.1) is the key this batch adds, with the same conditional the server itself applies: a non-empty string or no key at all — an empty string is neither a model id nor "unknown". The ledger is also checked against the fenced list in `docs/INTEGRATION-CLIENTS.md` §3d, so a doc that still says seven keys while the code forwards eight is red rather than merely stale. (2) The decide-failure arms carry the server's S-02 `currentPending` pointer key from a 409 `approval_stale` refusal onto the outcome the host reads. The reader is structural rather than `instanceof`, because the client is host-injected and the class identity is not this package's to assume; a half triple never mints (half a pointer cannot relocate anything), an empty string is not presence, and `checkpointToken` never transits. Both the allow and the deny leg are driven end to end through the real durable approval path — as is the accept-session leg, where a refusal carrying the pointer key must now re-raise instead of silently re-sending the human's answer for the **old** card as a plain approve (one decide call, pointer preserved), while a legacy 400 still falls back exactly as before — and all three flattening points must call the one shared reader — the same-shape residue check that makes "fixed one arm and left the twin" red instead of invisible. (3) The same disease growing on the REQUEST side: the `.mcp.json` → server-spec projection rebuilds each server key by key, and the settings schema deliberately leaves some keys parse-transparent — whatever JSON the file carries reaches the engine untouched, because validating them where the whole domain parses all-or-nothing would let one bad declaration take every server down silently. The whitelist had no row for the newest of them, so an operator's per-tool declarations — the ones the write fence reads — were stripped at the package boundary while both sides looked correct. The criterion is not "is that key handled" but the transparent-key table read out of the INSTALLED schema at runtime, reconciled name-for-name against this leg's ledger, so the day upstream adds a third one this turns red and forces an explicit decision. Behaviour is pinned on both transports, by object identity rather than deep equality (a rebuild would be a second judge), and malformed values must transit UNCHANGED rather than be refused here — the engine refuses them loudly and names the server, whereas a package-side judge can only swallow a declared protection quietly. Absence still mints no key, unknown keys still never reach the wire (the fix is the dropped key, not the gate), and the one transparent key this leg deliberately does not forward is a ledger entry with its own exit condition: it belongs to the deployment plane, and the day the request-plane type declares it the entry's premise is gone and the gate says so The prose probe reason on approval asks (`probeReason`: live frames from server 7.115.0, and the durable pending row's risk descriptor) lands on the same card field from both legs, byte for byte, never as an empty string, read from own keys only. A display field that throws when read (a getter or a proxy trap) is treated as absent: the card is still shown and the decision is still sent. The remember-refused flag (`rememberRefused`, the server saying an allow-for-this-session answer will not be remembered on this card) reaches the same card field from the live frame and from the pending-list row, only as an own strictly-true key; absent, malformed, inherited or throwing values leave it off without blocking the card. |
+| `scripts/run-additive-key-passthrough-test.mjs` | The one disease shape behind two legs: a **closed whitelist / flattening arm** dropping a fact that is already on the wire, while both sides of the seam look correct. (1) The `task_progress` projection carries a registered **key ledger** — a frame populated with every key the service really projects is pushed through the shipped `eventToSdkMessage`, and the set of wire keys that survive must equal the registered pass-through list **name for name in both directions**, so quietly forwarding one more key is as red as quietly dropping one. `model` (the child run's model id, minted by core as `prepared.model.id` and projected by the server since 7.52.1) is the key this batch adds, with the same conditional the server itself applies: a non-empty string or no key at all — an empty string is neither a model id nor "unknown". The ledger is also checked against the fenced list in `docs/INTEGRATION-CLIENTS.md` §3d, so a doc that still says seven keys while the code forwards eight is red rather than merely stale. (2) The decide-failure arms carry the server's S-02 `currentPending` pointer key from a 409 `approval_stale` refusal onto the outcome the host reads. The reader is structural rather than `instanceof`, because the client is host-injected and the class identity is not this package's to assume; a half triple never mints (half a pointer cannot relocate anything), an empty string is not presence, and `checkpointToken` never transits. Both the allow and the deny leg are driven end to end through the real durable approval path — as is the accept-session leg, where a refusal carrying the pointer key must now re-raise instead of silently re-sending the human's answer for the **old** card as a plain approve (one decide call, pointer preserved), while a legacy 400 still falls back exactly as before — and all three flattening points must call the one shared reader — the same-shape residue check that makes "fixed one arm and left the twin" red instead of invisible. (3) The same disease growing on the REQUEST side: the `.mcp.json` → server-spec projection rebuilds each server key by key, and the settings schema deliberately leaves some keys parse-transparent — whatever JSON the file carries reaches the engine untouched, because validating them where the whole domain parses all-or-nothing would let one bad declaration take every server down silently. The whitelist had no row for the newest of them, so an operator's per-tool declarations — the ones the write fence reads — were stripped at the package boundary while both sides looked correct. The criterion is not "is that key handled" but the transparent-key table read out of the INSTALLED schema at runtime, reconciled name-for-name against this leg's ledger, so the day upstream adds a third one this turns red and forces an explicit decision. Behaviour is pinned on both transports, by object identity rather than deep equality (a rebuild would be a second judge), and malformed values must transit UNCHANGED rather than be refused here — the engine refuses them loudly and names the server, whereas a package-side judge can only swallow a declared protection quietly. Absence still mints no key, unknown keys still never reach the wire (the fix is the dropped key, not the gate), and the one transparent key this leg deliberately does not forward is a ledger entry with its own exit condition: it belongs to the deployment plane, and the day the request-plane type declares it the entry's premise is gone and the gate says so The prose probe reason on approval asks (`probeReason`: live frames from server 7.115.0, and the durable pending row's risk descriptor) lands on the same card field from both legs, byte for byte, never as an empty string, read from own keys only. A display field that throws when read (a getter or a proxy trap) is treated as absent: the card is still shown and the decision is still sent. The remember-refused flag (`rememberRefused`, the server saying an allow-for-this-session answer will not be remembered on this card) reaches the same card field from the live frame and from the pending-list row, only as an own strictly-true key; absent, malformed, inherited or throwing values leave it off without blocking the card. The single reader of the card’s remember-refused declaration is a root export — an own key holding exactly `true`, never throwing on hostile input, agreeing with both card legs over the same pool — and no private copy remains. |
 | `scripts/run-esc-halt-plan-test.mjs` | The Esc stop decision every client shares: fire the **turn-level** halt first, and escalate to a **run-level** cancel in exactly two cases — the engine itself answered with a 409 from the closed code set (it is saying "there is no in-flight turn here; use cancel for a run-level stop"), or that shot came back with no verdict at all *and* the shell can independently prove a permission card was on screen. Everything else does not escalate. The asymmetry is the whole point and every negative control guards the same direction — deciding *not* to escalate costs the user one more choice on a busy-session card (recoverable), deciding to escalate wrongly tears down a run that was alive and takes every in-flight tool with it (not). So: the closed code set is a **frozen** value, not a `ReadonlySet` — type-level immutability does not stop a consumer's `.add()`, and the guard proves it by really trying to mutate the exported value and then checking the verdict did not drift; the escalation gate is the **conjunction** of that closed set and the 409 status, since honouring the code alone lets a 500 that merely quotes it drive a destructive call; `interrupt.not_held` and `steering.not_running` are deliberately outside the set (the first means *this replica* has no live face — the run may be perfectly alive on another); an unreadable code falls to the no-escalation side; a `parked` flag never overrides a verdict the engine did give, and only strict `true` counts when it did not. The first shot is unconditional by construction — it does not consult `parked`, because the 409 it earns is exactly the verdict the gate wants — and the verdict itself is a closed machine-readable reason word, not display copy. A third escalating case was added once tearing the stream stopped reaping the run: with detach armed, a shot that never lands leaves the run going all the way to the end of the turn, so the Esc the user pressed has no effect at all and nothing on screen says so — the old behaviour had a silent backstop (tearing the stream ended the run) and that backstop is gone. The new fact is held to the same three disciplines as `parked`: it is read only where the engine gave no verdict, it is judged **after** `parked` so an existing host's reason word does not change under it, and only strict `true` counts. Absence is proven to be a no-op rather than asserted — the guard carries its own reference implementation of the previous version's table, runs the full grid through both, requires zero divergence when the new field is omitted, and first shows the comparison really does report a difference on the one cell where the two versions are meant to differ |
 | `scripts/run-peer-frame-projection-test.mjs` | The three engine-injected lanes design/385 puts on the **one** `task_notification` carrier, which are not the same kind of thing at all: a delegated child's uplink (`agentMessage`), another session's message drained from this session's own box (`crossSessionMessage`), and a receipt about one of *this* session's own outbound messages (`crossSessionNotice`). The engine renders none of them inside a `<task-notification>` shell, so a client that projects them as the generic completion card shows "background task finished" while the model read a colleague's sentence — two faces describing different events. The discriminator is pinned to the **typed carrier being present**, never to the `summary` text: those carriers can only be minted by the engine's injection legs (the external `notify()` input is a strict subset of the payload and can wear none of them), while `summary` is filled by every notification there is — so anchoring on text would let any background task impersonate a colleague's message by writing `<agent-message from="…">` into its own summary, and a positive control asserts exactly that payload still projects as the generic card. Fail-closed has two tiers rather than one: a broken **required** field (empty `from`, a non-string `body`, a notice `kind` outside the closed set) returns absence so the caller falls back to the generic card — an honest downgrade where the user still sees the notification — while a broken **optional** field drops only itself, because losing an attribution note and losing a colleague's whole message are not the same magnitude. The provenance side record is **required and must agree on four points** (`kind` matches the lane; `from`/`taskId`/`seq` are present and equal the carrier/payload — each equality is anchored on a core mint site and pinned by the cli wire-anchor A-K24), so a carrier signed with a trusted name but a disagreeing provenance falls back to the generic card; peer bodies pass the same authority-envelope neutralization core applies (`<task-notification>` etc. are defused) so a colleague's text can never seed the resume dedup ledger. Lane precedence copies the engine renderer's own order, because the model already read the frame in that order and a client ordering of its own would put a card on screen that disagrees with the frame the model saw. Rendering and parsing of the transcript line live in the same module and are round-tripped in both directions, including a body carrying a forged closing tag (a parser fooled there hands half a message to the next row) and a quote inside the sender label (which must not forge a second attribute); the notice lane is deliberately kept **out** of the parser, since recognising it would mean anchoring the `[Cross-session …]` prefix and a user typing that same line would be rendered as engine speech. Hostile carriers are read as own **data** descriptors only and accessors are never invoked at all — `catch` catches throwing, not never returning — proven by a counting getter that must stay at zero calls, alongside a revoked proxy and a prototype-only carrier; and four legacy payload shapes assert the no-carrier path is byte-identical to before, which is the executable form of "zero difference for an older host". A re-supplied cross-session message — same task id, status and sequence as the first delivery, handed to the model again after compaction — is not rendered a second time, because the first delivery is still on the user's screen; a record with the next sequence number still renders |
 | `scripts/run-wiring-manifest-projection-test.mjs` | The two end-user facts carried on the engine's `wiring_manifest` frame (`modelGate`: which tools this run's model gate removed and the verbatim restore hint; `autoMode`: whether auto mode is actually armed and the engine's own reason word). Projection: both sections ride as `_sema_`-prefixed superset keys, verbatim, and no SDK-named key is minted; a frame where neither section is well-formed projects to `none/not_in_slice` (no empty arm); `modelGate` needs all three keys and treats `removed: []` as a bad value rather than a reading; `autoMode` needs a boolean plus a non-empty reason that agrees with it, and the reason word is never mapped onto the capabilities vocabulary; the frame is flat (a nested `manifest:{}` wrapper is not a supply); `eventId` rides like every other arm. Adapter: exactly one chrome event on the main lane, a sub-flow frame (any `parentToolCallId`, `null` included) yields nothing, and an absent `eventId` leaves the key absent. Added at receiving time because the shell-side gate could not see this package's behaviour: two mutations (empty `removed` accepted, sub-flow gate removed) had passed the package suite untouched 0.71.0 adds sections F–I: the fourth/fifth/sixth manifest sections (`tools` via the roster reader, `hooks[]` rows dropped one by one when malformed, `lsp` absent unless `mounted` is a boolean), the `tool_roster_delta` arm (narrowed `delta`, `malformed` when `fromDigest`/`roster` cannot be read, host applies it against its own digest), the `context_usage` arm (finite-gated scalars plus `sections[]` rows dropped one by one), and the `WiringManifestMcpEntryView` rename with `MAX_AGENT_SKILLS` gone from the surface From 0.84.0 the `hands` section (whether this leg was assembled with the engine's built-in file and shell tools) is projected as well: a strict boolean `mounted` plus the engine's reason word passed through as written, absence kept as "not reported" rather than folded to either answer; a three-state reader and one sentence source share their opening and closing with the roster-derived sentence, whose bytes do not change. The sentence source answers the not-reported sentence for anything it does not recognise, including a roster-derived reading, and never throws. From 0.88.0 the `fileHistory` section (whether this leg records file history; `on` / `disabled` today, passed through as written for unknown words) is projected as well: present only as a non-empty own string, absence kept as "not reported" rather than folded to `disabled`, malformed values dropped like the server drops them, the same view on the non-streaming submit receipt, and a leg built and published by an installed server package checked end to end when one is available. A `fileHistory` value that throws when read leaves only that section absent; the other sections of the same frame are still delivered on both the streaming and the submit-receipt paths. The server-minted `instructions` section (which instruction file the leg loaded) is projected verbatim as a frozen copy, an explicit empty selection stays an empty list, absence stays absent, a malformed section is dropped without touching its siblings, and a frame carrying only this section still becomes an event. The `fleet.subagentTranscripts` statement (from server 7.115.0: how long delegated sub-agent conversations survive) is passed through verbatim as an open word in a frozen `fleet` section; absence stays absent and is never read as `none`, and a malformed value drops only that section. |
@@ -388,28 +388,28 @@ guard still cross-checks the table by name).
 | `scripts/run-model-capability-probe-test.mjs` | Whether a model on the OpenAI-completions lane **thinks**, and whether that thinking can be **turned off** — a question nobody can answer by looking at a model name, and one whose wrong answer costs every later call. The probe is judgement only: the network half arrives as an injected port, so the package mints no URL, reads no credential and never calls `fetch` — pinned by a source-level assertion, because a package that reaches the network once has changed what every host must trust it with. The seven dialect words are a **copy**, reconciled element-wise against the installed engine’s own bytes in both directions, since the words belong upstream and a private table drifts the day a dialect is added; the settings package deliberately declines to restate them, so the table cannot be imported from there and this guard is what stands in for the import. The **order** the dialects are tried in is a public promise rather than an implementation detail — each extra attempt is real money and real latency against someone’s gateway — so the guard pins the exact call sequence a stub records, and reversing it reds on the wasted round trip; the template-parameter spelling leads because an observed gateway keeps thinking, and answers with an empty body, when handed the top-level switch instead. That observation is also why an empty answer is **not** accepted as *thinking is off*: a knob that deletes the reply is not a knob that disabled reasoning, and accepting it would write a spelling into the catalogue that the gateway does not honour. Two dialect words whose request bytes are identical to another’s do not each burn an attempt. The two verdicts that look alike are held apart from both directions: *tried everything, still thinking* requires at least one attempt to have **cleanly answered**, and when every attempt was refused the verdict is *could not tell* instead — and on the unanswerable path the result carries **no** thinking flag at all rather than a fabricated `false`, while the pure write-back returns the very same entry object untouched. A verdict that reasoning cannot be disabled **removes** a previously declared spelling rather than leaving it, since a refuted spelling keeps the engine sending bytes the gateway ignores while the catalogue still renders it as already off. Evidence is lengths, finish positions and status codes only — a planted secret in both the answer and the reasoning channel must appear nowhere in the result, so the record can go into a log or a ticket whole. One cross-package premise is checked by really running the other package’s parser rather than quoting its documentation: everything this probe writes eventually passes through the settings schema on its way into a catalogue, and that field is declared parse-transparent precisely so the vocabulary can live on the consuming side. If it ever narrows, the spelling is stripped **silently** — indistinguishable from the probe never having run — so the guard feeds the probe’s real output through the real parser, checks the compat object comes back key for key, and checks a dialect word this client has never heard of survives too. Two shapes that must be rejected really are rejected, since otherwise the survival checks would hold on a parser that accepts anything, and a bare entry is asserted valid first, because the first run of this section reddened on a space in a fixture’s name — a fixture that cannot pass would disguise the real alarm as already having fired |
 | `scripts/run-decide-receipt-test.mjs` | What a decision verb actually **answered** — and, more importantly, what it did not. A success response on the newest lane is only an acknowledgement that the decision was accepted for delivery: the approval is still pending, and a client that clears the card on it shows either a ghost card that was already approved or a card that vanished while the decision was lost. So the package deliberately has **no** "was it resolved" predicate — nothing in that body can answer it — only the opposite one, whose `false` is likewise not evidence of resolution; resolution is only ever the next running arm on the stream. The guard pins that inversion in the product source too: the success path must no longer clear the latched gate, while the stream-observing path that really clears it must still be there. The body has four shapes with **no** key common to all of them, so every position is read as honestly absent, and the handoff handle — which run to watch from here on — requires **two** facts together, since either one alone would either point the stream at the run it already had or mint an empty handle. The record of what finally happened to an already-decided action is read through the **same** reader as every other gate record rather than a second copy, and its absence means **unknown**, never *it was allowed* — the two can even contradict each other, so the card says nothing at all when it is missing. The three refusals on that lane each get one distinct sentence and a disposition taken from **why** each was refused rather than from severity: one cannot be helped by re-sending at all, one waits on the host, one just drops an option — and none of them carries a countdown, because the server never mints a wait for them. Recognition is a **closed set**: an unrecognised code on the same prefix returns nothing rather than a guess, since that prefix also houses a safety signal whose whole rule is never to retry automatically, and the recovery handle is read as absent when unreadable rather than substituted from a different identifier that no longer appears on that lane **0.68.3 (core 7.18.0):** `gate.disposition.classifier` is read key by key into a named view (requested model, model that answered, and the ladder fallback when one happened); a half-shaped record yields no classifier at all rather than a half view, and absence stays "unknown", never "the seat answered itself" |
 | `scripts/run-approval-frame-chrome-arms-test.mjs` | The two in-stream approval frames finally reaching every host through the shared pipeline instead of one shell's private branch — the shape of a layering defect: hosts that only consume the package could not rebuild their pending cards after a reconnect, and did not clear a card the engine had withdrawn. The payload is deliberately carried as the **envelope** the upstream types declare rather than the first-version card: the stream parser applies no predicate, so narrowing here would let a legitimately newer frame pass as the older shape and invite consumers to read keys a newer card never promised. The guard therefore pins that every open key survives untouched, that an unknown version still passes through, and that narrowing is left to the host's own predicates — with the fallback being a generic card and a person, **never** an automatic denial. A frame whose version cannot be read at all is reported as malformed rather than dropped in silence, because both frames carry user-visible decisions and state changes. Both arms are registered as **required** host duties, and their duty text names the load-bearing rules a host would otherwise have to rediscover: which predicate to narrow with, that the reconnect preamble — not a replayed historical frame — is the authority on which cards exist, and that a withdrawal frame can be lost entirely. Unlike the sibling arms, these carry **no** sub-stream cutoff: an approval raised under a delegated call still has to reach a person, and filtering it by ownership is the host's job, not a reason to discard it. Finally the upstream bytes that justify the envelope discipline are checked to still be there, since the whole design rests on them |
-| `scripts/run-terminal-status-vocabulary-test.mjs` | One place that decides whether a run has **ended** and whether it ended badly — written because that judgement had already been hand-copied three times, so the day the engine added a word for *the agent itself reported it cannot continue*, every copy missed it and a panel settled a self-reported failure as a success. The distinction the table exists for is pinned from both sides: that word belongs in it, while the two words meaning *waiting for a person to decide* deliberately do **not** — reading those as endings would bury a run that is actively waiting on the reader. A word this client does not know answers *no*, and the guard states plainly that *no* is not evidence of success: proving success means reading the positive side, so negating this predicate is the very mistake that caused two earlier incidents. The fleet lane gets the same treatment from the other direction: a workflow parked on a durable approval used to fall through to *running*, leaving the person with no hint that a card was waiting, and it now lands on the same rendered word the task lane already used — same fact, same word, checked end to end on a real row. Why the word was added directly rather than carried as a private superset key is checked mechanically against the upstream declaration being open, so the day it closes this reds and the decision gets revisited. The residue sweep is the point: the source tree must contain **no** further inlined copy of the judgement, each of the three former sites is checked to really read the single predicate, and the one reviewed exemption carries its reason **and** a liveness assertion, so an exemption whose justification expires cannot quietly keep standing |
-| `scripts/run-terminal-word-source-test.mjs` | Two tables of ending words, kept apart by **who owns them** — because they used to be one. The engine's own closed set of reasons a run ended, and the server's set of row states a run can finish in, overlap in three words but not in all of them: one word for *something outside stopped it* exists only on the server side, and one for *it paused and can be resumed* exists only on the engine side and means very nearly the opposite of an ending. Merged into a single list, those two sources became indistinguishable, so a new word on either side looked the same as a new word on the other, and the safest-looking move — folding the unknown word into a known one — is the exact mistake that has caused incidents here before. The engine-owned table is checked as a **copy, not an opinion**: it is reconciled word-for-word and in order against the installed engine package, read from both its declaration and its runtime bytes with the two required to agree, so the day upstream adds a fifth reason this reds before anything ships. The two dividing words are each pinned from both sides, including against the upstream declaration directly rather than only against this package's own list. Why the table was copied rather than re-exported was itself an assertion with an expiry, and it has expired: the engine package now publishes the set as a value, and since the package still may not import the engine at run time, the table is generated at build time from that published value — the guard checks that the package's table is that generated object and matches the engine's published keys in order. The renamed tables leave **no alias** behind, since an alias would let a reader keep consuming the merged list and the split would have bought nothing |
-| `scripts/run-terminal-table-provenance-test.mjs` | Several tables answering *has this ended*, which until now only asserted their own current wording rather than that the wording was right — a snapshot equality passes forever even the day upstream adds a word this package never learns about. Each is reconciled against a named upstream source instead, one comparator shared across all of them rather than one copy per table: a notification's terminal words are the engine's own closed set minus its one live word; a sub-agent tick's terminal words are the engine's own inline status literal minus *running*; a run row's terminal words must **cover every** engine reason a run can end — missing one is the exact failure mode that once let a client retry a connection until its budget ran out while the ending sat unread in the row the whole time — plus one explicitly named legacy word the engine's current declaration no longer carries — kept on purpose to read rows stored before that word was retired, since retiring a word the engine writes does not remove rows already stored with it; a fleet row's terminal words are pinned to **exact equality** with the transport's own status set minus its known non-terminal words (an adversarial pass found the earlier one-directional form let a real terminal word be quietly deleted from this side and still pass), and separately pinned against that set's current member count so the day it changes a person has to look. A sixth table, a workflow's terminal words, has two upstream producers and is pinned to **exact equality** with their union minus the one live word: the stored run record's status set, and the statuses the in-process task registry gives a workflow handle (read from the engine's own source; it emits *cancelled* when a workflow without a store is stopped — a version of this table reconciled against the first producer alone dropped that word, so a stopped workflow was never marked delivered and its completion could be delivered twice). The table is separately checked against every non-terminal word gathered — including one meaning *durably paused*, sourced from the engine's own outcome vocabulary rather than any of the other unions, after the same adversarial pass found a caller-documented non-terminal word this boundary had missed — and a behavioral regression drives the installed engine's real task registry through *stop* then *read the output*, feeds that result through both projection paths, and requires each to mark the run delivered and clear it from the waiting count. A seventh pair, found by the same pass sweeping the whole tree for the same shape of hand-copied table, answers a related but distinct question — whether a session's claim on a run has been released or is still held — and is pinned as two complementary halves of one upstream set: released-minus-one-named-legacy-word and held must partition the transport's status set exactly, so a real state going missing from either side is caught the same way a fabricated one would be. Every extraction in this guard parses real syntax rather than pattern-matching quoted text, so a comment mentioning a word never counts as that word being present, and single- and double-quoted members are read identically |
+| `scripts/run-terminal-status-vocabulary-test.mjs` | One place that decides whether a run has **ended** and whether it ended badly — written because that judgement had already been hand-copied three times, so the day the engine added a word for *the agent itself reported it cannot continue*, every copy missed it and a panel settled a self-reported failure as a success. The distinction the table exists for is pinned from both sides: that word belongs in it, while the two words meaning *waiting for a person to decide* deliberately do **not** — reading those as endings would bury a run that is actively waiting on the reader. A word this client does not know answers *no*, and the guard states plainly that *no* is not evidence of success: proving success means reading the positive side, so negating this predicate is the very mistake that caused two earlier incidents. The fleet lane gets the same treatment from the other direction: a workflow parked on a durable approval used to fall through to *running*, leaving the person with no hint that a card was waiting, and it now lands on the same rendered word the task lane already used — same fact, same word, checked end to end on a real row. Why the word was added directly rather than carried as a private superset key is checked mechanically against the upstream declaration being open, so the day it closes this reds and the decision gets revisited. The residue sweep is the point: the source tree must contain **no** further inlined copy of the judgement, each of the three former sites is checked to really read the single predicate, and the exemption table is empty: every former inlined copy now reads the single predicate, so a new inlined copy turns the gate red instead of being waved through. Comments are stripped with the shared token-level reader rather than a hand-written quote-pairing scanner, so a regular-expression literal that merely contains a comment opener can no longer swallow the code after it; a planted input pins that from both sides |
+| `scripts/run-terminal-word-source-test.mjs` | The dividing lines between the owner-named terminal tables, checked from the public entry point. The word for *something outside stopped it* is not an engine reason and not a run-row ending; it appears only in the fleet, notification and task-registry tables, so reading it through the run-row table is a type error rather than a silent miss. The word for *paused and resumable* is an engine reason but appears in no terminal table, so it can never be read as an ending. The word for *the agent reported it cannot continue* is a run-row ending only, and *cancelled* belongs to the notification and task-registry tables only. Every public table is a frozen literal tuple a consumer can derive types from; every retired name — the two from the first split and the ten renamed when owners were added to the names — is absent rather than aliased; internal-only tables and predicates stay off the public surface. Every predicate tolerates malformed input, answers *no* for unknown words and prototype-chain names, and the guard states plainly that *no* is not evidence of success. The run-terminal reader keeps one arm per engine reason, and an unknown reason falls to the unknown arm with the word carried back verbatim. Word-for-word reconciliation against the upstream packages lives in the parameterized owner guard. |
+| `scripts/run-terminal-table-provenance-test.mjs` | Every table in this package that answers *has this ended* is a machine-read copy of the terminal half of one named upstream union, lives in one owner module, and is named after its owner — and one parameterized guard holds all of them to that rule. Each registered row pairs a table with the upstream type it copies, the explicitly listed non-terminal half of that union, and any named legacy word kept only to read rows stored before the word was retired; a single comparator then requires the table plus the non-terminal half to equal the upstream union exactly, in both directions. Both sides are resolved by the TypeScript checker from the installed packages' public entry points rather than by pattern-matching declaration text, so a word added upstream, a word dropped locally, a word invented locally, an overlap between the two halves, a legacy word that is still live upstream, or an open-ended member each turn the guard red — and planted fakes prove each of those shapes is caught before the real rows are judged. The guard also checks the naming rule mechanically, finds no terminal table declared and no membership test against one anywhere outside the owner module — independent of names: a table may be imported only at registered sites (any alias, namespace, dynamic import or re-export counts), and no string array literal or equality chain built from two or more distinct terminal words may appear outside the owner module — requires each converted read point to call its owner's predicate, checks the shipped values are frozen and equal to their declared member types, checks the projection onto the local task-row words, and replays two behaviours that changed when read points moved to their true owner: a cancelled notification now settles its row as an error and announces the terminal frame, and a cancelled background child is no longer settled as a success. A behavioural regression drives the installed engine's real task registry through stop-then-read and requires both projection paths to mark the run delivered. |
 | `scripts/run-workflow-park-truth-projection-test.mjs` | The read face for *which approvals a workflow run left parked* — and the credential that must never ride along with it. Upstream strips the redemption token from that response, and this package's reader is built so the token **cannot** come back: each row is assembled field by field from the three identity keys, never copied wholesale, so an extra key appearing upstream is structurally unable to reach anything this package hands a UI. The guard proves that rather than asserting it — a poisoned row carrying a secret is read, and the secret is searched for across the **entire** serialized result, with the same search proven to find it in the input so a blind search cannot pass; renaming the credential key does not help it through, because the rule is *only these three*, not a blocklist; and the reader's own source is checked to contain no object spread, since one such line would quietly void all of it. The other half is an absence distinction with opposite consequences: a record with **no** parks field at all was written by an older engine and proves nothing about whether approvals are waiting, while an empty list is a positive statement that none are — collapsing those two would let a run whose parked approvals cannot be proven be resumed anyway, so they are kept literally distinguishable, and a payload whose rows are all unreadable answers *unknown* rather than *none*. The four refusal codes for this family are checked code by code against the engine's real bytes, never matched by name prefix, and the older umbrella code they were split out of is asserted to still be **alive** — treating the whole code as retired would make a family of real refusals vanish silently **0.68.3 (core 7.18.0):** two more keys ride the same projection duty as `parks` itself: `originUnconfirmed: true` on a row (never `false`; absence is the confirmed state) and `resumeAdmissionIncomplete: true` on the run (presence means "not a resume base"). Dropping either would turn a refused record back into an admissible one, so the guard pins both, including that neither folds into the other **0.69.1 (CC-12):** both keys now also ride the projected `WorkflowRunState`, so a host that only sees the projection can render them |
 | `scripts/run-retired-vocabulary-census-test.mjs` | Whether a retirement really happened. When upstream removes a family, a downstream package can cut it out or keep a courteous alias — and the alias is the worse outcome: three clients keep writing branches for something nobody emits, and a status line advertises a state it can never reach. Choosing the clean cut only means something if a guard holds it, since a comment saying *retired* is not an exit code. Each registered entry is held two ways: the name must be gone from **code positions** in this package (comments stripped first, because the explanation is supposed to stay) and off the published surface, and — the half that keeps this from being self-congratulation — it must really be gone **upstream**, since that is the entire reason it was removed here; if it comes back, the disposition deserves reconsideration rather than silence. The scanner proves it can speak by finding a symbol that is genuinely present before any absence is believed, and distinguishes a mention inside a comment from one in a string literal, which is exactly the form being cleared. A closing check runs the other way: the retirement **story** must remain in the comments, including a promise this package made earlier and has now had to withdraw — deleting the history alongside the code is a bad way to satisfy *zero hits*, and leaves the next reader with code that has no reason |
-| `scripts/run-classifier-status-test.mjs` | What state the auto-mode classifier is in **on this session** — the question a doctor line, a model settings page and a permission card’s status row all ask, and a different question from the one the approval card asks (*why am I being asked right now*), so the sentences are pinned mutually distinct from that face’s as well as from each other. The session-level half of this reading — a breaker record the engine used to keep — was **retired upstream**, and the guard now holds that retirement from **both** sides: the engine's own declarations must really no longer carry it (a fact coming back would mean the removal here was the wrong disposition, and that deserves a conversation rather than silence), and this package must carry no alias, no state word and no leftover narrowing for it — a reading kept alive for something nobody emits any more is a promise the interface cannot keep, and it left the doctor line advertising a state it can never reach. What remains is ordered by the quantity that actually decides whether the classifier is running: the fact from **this round** first, then whether this leg is armed — a decider is minted per run, so a later leg can be armed again. Not armed, and a section that never arrived, both answer **undefined** rather than *available*; that arming question has its own field and answering it twice grows a second ledger. Arming and availability are also **two words, not one**: the engine says a decider was minted *for this leg*, which is an assembly-time fact, while whether that decider answers any given round is a **per-call** one — so an armed leg reads `armed` and only a positive per-call fact (an ask whose origin is the classifier's own denial-bound fallback, which by construction stands *after* the classifier ran) reads `available`. Every other ask origin is refused as evidence and for a stated reason rather than out of caution: several are ones the classifier is structurally forbidden to answer, and for the rest a surviving ask is precisely the case where it did **not** resolve one — so reading availability off them would be a guess. The projection is a **whitelist**, so an older engine still sending the retired member loses it at the boundary while the two live facts beside it ride through untouched. Rendering never throws and never impersonates: a state word this client does not know — including the retired one, which a restored view can still carry — reaches an honest fallback that names it verbatim, carries no invented explanation of a mechanism that no longer exists, and is proven distinct from all three real sentences; prototype keys reach that same fallback rather than a function body, checked against a real out-of-table word so the comparison cannot hold vacuously |
+| `scripts/run-classifier-status-test.mjs` | What state the auto-mode classifier is in **on this session** — the question a doctor line, a model settings page and a permission card’s status row all ask, and a different question from the one the approval card asks (*why am I being asked right now*), so the sentences are pinned mutually distinct from that face’s as well as from each other. The session-level half of this reading — a breaker record the engine used to keep — was **retired upstream**, and the guard now holds that retirement from **both** sides: the engine's own declarations must really no longer carry it (a fact coming back would mean the removal here was the wrong disposition, and that deserves a conversation rather than silence), and this package must carry no alias, no state word and no leftover narrowing for it — a reading kept alive for something nobody emits any more is a promise the interface cannot keep, and it left the doctor line advertising a state it can never reach. What remains is ordered by the quantity that actually decides whether the classifier is running: the fact from **this round** first, then whether this leg is armed — a decider is minted per run, so a later leg can be armed again. Not armed, and a section that never arrived, both answer **undefined** rather than *available*; that arming question has its own field and answering it twice grows a second ledger. Arming and availability are also **two words, not one**: the engine says a decider was minted *for this leg*, which is an assembly-time fact, while whether that decider answers any given round is a **per-call** one — so an armed leg reads `armed` and only a positive per-call fact (an ask whose origin is the classifier's own denial-bound fallback, which by construction stands *after* the classifier ran) reads `available`. Every other ask origin is refused as evidence and for a stated reason rather than out of caution: several are ones the classifier is structurally forbidden to answer, and for the rest a surviving ask is precisely the case where it did **not** resolve one — so reading availability off them would be a guess. The projection is a **whitelist**, so an older engine still sending the retired member loses it at the boundary while the two live facts beside it ride through untouched. Rendering never throws and never impersonates: a state word this client does not know — including the retired one, which a restored view can still carry — reaches an honest fallback that names it verbatim, carries no invented explanation of a mechanism that no longer exists, and is proven distinct from all three real sentences; prototype keys reach that same fallback rather than a function body, checked against a real out-of-table word so the comparison cannot hold vacuously The ask origin is read the one way it is read everywhere else — an own key holding a non-empty string — so an origin that exists only on a prototype no longer counts as proof that the classifier ran. |
 | `scripts/run-compaction-boundary-projection-test.mjs` | The compaction divider and the one frame that makes its anchor resolvable. The trigger word is passed through as an **open set** instead of being folded to two: the engine deliberately stopped flattening its third value (a compaction that was not optional — a prompt-too-long recovery or trim pressure) and carries what the hook layer saw, so folding it again at the package boundary re-introduces exactly what upstream had just removed, while a consumer branching on *is it manual* keeps its behaviour byte for byte. Only an unreadable word (absent, empty, non-string) falls back — that is *could not read it*, not *read it and did not recognise it*. Two superset keys ride the metadata and neither fabricates: the preserved-segment anchor is minted only when its id really reads out, because half an anchor sends the host looking up an empty string in its map, and the clamp ratio is a **disclosure** whose real zero is a fact rather than an absence. The clamp ratio also carries a registered exit condition — the service really sends it while the SDK arm has no seat for it yet, so the read is defensive and this guard reds the day that seat appears, forcing a re-check instead of leaving a cast to rot. The committed-message frame moves out of *deliberately not projected*: that classification was true about transcript rows and false about **positioning**, since the engine states that consumers build their own id-to-message map from this frame to place the divider — projecting the anchor without it hands the host something it cannot resolve. It becomes a neutral internal arm and an optional chrome ledger event, never a transcript row (the frame carries no body, so minting one would put words in the engine's mouth), with both required ids narrowed and a malformed frame recorded rather than half-minted |
 | `scripts/run-cost-absence-projection-test.mjs` | Telling **declared free** apart from **never priced**, in both directions, because the package was getting each one wrong in the opposite way. The engine separates them on the wire — an absent cost means some spend had no price table, an explicit zero means the model declared itself free — and the result projector used to require a *positive* number, so a genuinely free run could not say so; while the per-model mirror folded absence to zero, so an unpriced run told a billing consumer it cost nothing. The total is now reported as the engine stated it, with absence and non-finite values alone reading as unknown, and a negative passed through rather than corrected, since a refund is a legal figure and the package is not a second accountant. The per-model figure keeps the CC shape intact — that field is a required number and *unknown* is simply not expressible in it — so the value stays zero and a **companion superset bit** carries the distinction, which means the two are read together and a reader that only ever looked at the number is unchanged; the bit is minted only in the absent case and never as `false`, since a key present with a false value reads as a third state. The same mint point serves both the wire's per-model split and the synthesised current-model row, so neither can drift. Alongside it the cache-write figure stops being a hardcoded zero and reads the field the wire has always carried, in both the flat usage and the synthesised row, and all four flat token slots move from a null-coalesce to a finite-number guard — the stats object has an open index signature and the wire is JSON, so a string or an infinity would otherwise land in a slot the types promise is a number, compiling green and surfacing only when something sums it |
-| `scripts/run-permission-denial-projection-test.mjs` | The terminal result's **permission-denial list** being the wire's real one rather than a hardcoded empty array. The session vocabulary carries a list of tool calls that were denied; the projector used to mint `[]` in both the success arm and the error envelope, which folded two different statements into one — *nothing was denied on this run* and *this frame carries no such ledger at all* (an older engine, a rejection envelope, a failure event that arrives without stats) looked identical. Each denied gate on the wire's human-review ledger now becomes one record, in wire order, carrying the keys the wire can actually honour: the tool name when it reported one, and a superset field with the engine's own short, redacted one-line summary of the call's input. **Two lists, deliberately.** The reference shape requires three fields on every element — tool name, call id, and the full input object — and the wire's ledger carries only the first. Filling the other two with an empty string and an empty object would be invention; putting a half-filled element into the reference array would break the element contract, and a strict consumer validating the stream drops the *whole* result message rather than one field. So the reference array admits only fully-formed records — empty today, and filling itself the day the wire grows the two missing fields, with no code change — while every record the wire really has rides a superset carrier beside it. A contract check pins today's absence, so that day turns this guard red on purpose. The companion bit means *this reference list cannot be claimed complete*: no ledger, an unreadable row, an unrecognised decision word (a rejected plan is not a denied tool call, and a row with no decision at all is not a judgement), or a record that could not be fully formed. Only its absence lets a reader say *zero denials*; it is never minted as `false`. Rows that cannot be read drop themselves rather than the whole ledger, and both arms go through one mint point so they cannot drift. Since 0.73.4 the third CC key is sourced from the same stream's `tool_start` frame, joined by call id: a row joins only when the frame was seen on this stream, its arguments are a plain object, and the scan over them finishes within budget. Since 0.84.1 arguments the transport redacted on the way (a string leaf carrying a replacement token, or a cycle / depth placeholder) join as well: the reference list receives the very object the transcript's tool call shows, and that entry, on both lists, carries a superset flag saying the input is a redacted view rather than the original — redacted arguments that are over budget, not a plain object, or never seen on this stream still stay off the reference list, and the flag is never minted as false; both halves have positive controls (a fully joined list drops the discriminator, a partially joined one keeps it), the ledger's own input wins when present, the snapshot is per-stream and capped with a one-way overflow latch, and an id seen with two different argument objects never joins. Later sections add the second stream-local join and the two discriminators the headless exit-code rule needs. "Which layer denied this" is not on the denial ledger at all — it is on the gate record of the same call's close-out frame, so it is joined by call id under the same law as the arguments: the closed word table is checked on the collecting side, the ledger's own value wins if it ever arrives, a word from outside the table is not stamped, and a row that cannot be joined keeps the key absent rather than claiming nobody denied it. The classification word is carried on both lists under the same name and the same value, so a consumer needs one reader, not two. The "this run produced no tool output and was denied" flag is present only when three independent things hold at once — the denial evidence is read from the full list rather than the strict one, which can be empty for reasons that have nothing to do with denials; this stream saw no successful tool close-out; and this stream can honestly claim to have watched the run from its first frame. A stream that reconnected mid-run cannot make the last claim, so it mints nothing rather than a false negative, and the flag is never minted as false. From 0.80.0 that classification has a **second source**. It used to come only from this package's own decision path, so a refusal the engine settled on its own — a deployment policy answering the card on an unattended lane, with no client involved — left the field empty even though the same stream's gate record said exactly what had happened. The engine's own settlement word now fills it when, and only when, the local one is absent: the package's own attribution always wins, because letting a replayed frame overwrite it would let the wire change what the host itself said. The word is read literally in both directions and never reverse-engineered, and the separate field naming *which layer* refused is left exactly as the wire wrote it — the two answer different questions, and rewriting one to match the other would make them say the same thing twice. A later section reconciles the terminal list against the denied calls seen in the same stream, so denials that never reached a human (rules, hooks, classifiers, write protection) are listed too: complete rows join the CC list, rows missing a field stay on the extended list and mark it incomplete. A further section feeds the same raw tool-result frame through the projector into both lanes and requires the denial category on the interactive transcript record, on the non-interactive frame and from the reader on the raw frame to agree, including frames whose settlement, classifier cause or classifier attribution is present but unreadable — a shape the narrowed gate view on the internal frame cannot show — and pins that the reader gives the same answer on the internal frame as on the raw one. |
+| `scripts/run-permission-denial-projection-test.mjs` | The terminal result's **permission-denial list** being the wire's real one rather than a hardcoded empty array. The session vocabulary carries a list of tool calls that were denied; the projector used to mint `[]` in both the success arm and the error envelope, which folded two different statements into one — *nothing was denied on this run* and *this frame carries no such ledger at all* (an older engine, a rejection envelope, a failure event that arrives without stats) looked identical. Each denied gate on the wire's human-review ledger now becomes one record, in wire order, carrying the keys the wire can actually honour: the tool name when it reported one, and a superset field with the engine's own short, redacted one-line summary of the call's input. **Two lists, deliberately.** The reference shape requires three fields on every element — tool name, call id, and the full input object — and the wire's ledger carries only the first. Filling the other two with an empty string and an empty object would be invention; putting a half-filled element into the reference array would break the element contract, and a strict consumer validating the stream drops the *whole* result message rather than one field. So the reference array admits only fully-formed records — empty today, and filling itself the day the wire grows the two missing fields, with no code change — while every record the wire really has rides a superset carrier beside it. A contract check pins today's absence, so that day turns this guard red on purpose. The companion bit means *this reference list cannot be claimed complete*: no ledger, an unreadable row, an unrecognised decision word (a rejected plan is not a denied tool call, and a row with no decision at all is not a judgement), or a record that could not be fully formed, or — since 0.95.0 — a stream that did not cover the whole run (no first frame seen on it, or no denied-call table handed over; absence of either reading means unknown, not covered). Only its absence lets a reader say *zero denials*; it is never minted as `false`. Rows that cannot be read drop themselves rather than the whole ledger, and both arms go through one mint point so they cannot drift. Since 0.73.4 the third CC key is sourced from the same stream's `tool_start` frame, joined by call id: a row joins only when the frame was seen on this stream, its arguments are a plain object, and the scan over them finishes within budget. Since 0.84.1 arguments the transport redacted on the way (a string leaf carrying a replacement token, or a cycle / depth placeholder) join as well: the reference list receives the very object the transcript's tool call shows, and that entry, on both lists, carries a superset flag saying the input is a redacted view rather than the original — redacted arguments that are over budget, not a plain object, or never seen on this stream still stay off the reference list, and the flag is never minted as false; both halves have positive controls (a fully joined list drops the discriminator, a partially joined one keeps it), the ledger's own input wins when present, the snapshot is per-stream and capped with a one-way overflow latch, and an id seen with two different argument objects never joins. Later sections add the second stream-local join and the two discriminators the headless exit-code rule needs. "Which layer denied this" is not on the denial ledger at all — it is on the gate record of the same call's close-out frame, so it is joined by call id under the same law as the arguments: the closed word table is checked on the collecting side, the ledger's own value wins if it ever arrives, a word from outside the table is not stamped, and a row that cannot be joined keeps the key absent rather than claiming nobody denied it. The classification word is carried on both lists under the same name and the same value, so a consumer needs one reader, not two. The "this run produced no tool output and was denied" flag is present only when three independent things hold at once — the denial evidence is read from the full list rather than the strict one, which can be empty for reasons that have nothing to do with denials; this stream saw no successful tool close-out; and this stream can honestly claim to have watched the run from its first frame. A stream that reconnected mid-run cannot make the last claim, so it mints nothing rather than a false negative, and the flag is never minted as false. From 0.80.0 that classification has a **second source**. It used to come only from this package's own decision path, so a refusal the engine settled on its own — a deployment policy answering the card on an unattended lane, with no client involved — left the field empty even though the same stream's gate record said exactly what had happened. The engine's own settlement word now fills it when, and only when, the local one is absent: the package's own attribution always wins, because letting a replayed frame overwrite it would let the wire change what the host itself said. The word is read literally in both directions and never reverse-engineered, and the separate field naming *which layer* refused is left exactly as the wire wrote it — the two answer different questions, and rewriting one to match the other would make them say the same thing twice. A call whose merged classification is one of the two abort words is listed on neither side, ledger row or stream-side addition (since 0.95.0 both read the same table, so a human refusal followed by a local interrupt is not reported as a denial); a row without an id, a missing classification or an unknown word is not guessed at and stays listed. A later section reconciles the terminal list against the denied calls seen in the same stream, so denials that never reached a human (rules, hooks, classifiers, write protection) are listed too: complete rows join the CC list, rows missing a field stay on the extended list and mark it incomplete. A further section feeds the same raw tool-result frame through the projector into both lanes and requires the denial category on the interactive transcript record, on the non-interactive frame and from the reader on the raw frame to agree, including frames whose settlement, classifier cause or classifier attribution is present but unreadable — a shape the narrowed gate view on the internal frame cannot show — and pins that the reader gives the same answer on the internal frame as on the raw one. The gate-record reader reads each property of a raw frame once, so a part’s value and its presence come from the same snapshot (a frame whose getter answers differently on a second read gets one consistent answer), and over a pool of several thousand plain-data frames both the view and the denial-class reader agree with the previous implementation frame for frame. |
 | `scripts/run-cost-reconcile-projection-test.mjs` | The **end-of-run cost reconciliation** reaching consumers at all. The engine splits a run's spend on the wire — the task's own cost, which deliberately excludes delegated sub-agents, the delegated total itself, and the within-task compaction subtotal that sits inside the own figure — and states two reconciliation identities for them. The package used to project none of it, so a cost view could only ever see one number and under-reported both delegated and compaction spend. Both structures are now projected onto the result as superset fields in the wire's integer micro-currency unit, read key by key, with unreadable keys dropped individually, an entirely unreadable structure omitted rather than emitted empty, and unknown categories passed through since the vocabulary belongs upstream. The delegated cost stays **absent when it was never priced**, never a fabricated zero. The same reader also feeds a terminal chrome arm carrying the three parts plus the reconciled total, so the two faces can never compute different answers; the reconciled total is minted only when both sides are known, and otherwise a discriminator bit says which side is unknown. **The reference field for total cost keeps its meaning** — it remains the task's own spend and the delegated total is not folded into it — because that is a shape the wider ecosystem reads; the reconciled figure is offered beside it, not in place of it. A frame that carries no stats emits no arm at all, and the existing rule that in-stream per-turn usage is not published for sub-flows is pinned unchanged, since delegated spend arrives once, at the end. The bit that says those figures are a lower bound is **per stream**, not per context: the emit context belongs to the caller and may be reused across streams, so a gap observed on one run is no evidence at all about the next one — the observation is held for the duration of one stream and handed to both projection faces by value, and the guard drives a reused context both sequentially and concurrently to prove neither direction leaks |
 | `scripts/run-task-progress-terminal-projection-test.mjs` | The one tick that says a delegated child **finished**. The engine fires exactly one final beat carrying a terminal face, and says in the same breath why it exists — so a consumer sees the row finish instead of watching it vanish after the last running beat — but the package's projection whitelist had no seat for that field and its adapter still carried the older premise in a comment, so the terminal beat arrived byte-identical to another running one: the panel row stayed up waiting for a defensive sweep (which only ever settles rows bound to a card still open this turn) or for a separate notification frame. The status now rides through as an **open set** with the vocabulary left upstream, while the question *which words are terminal* is answered by a closed pair on the adapter side — an unrecognised new word takes the running path, because guessing it terminal ends a row that is still working whereas one extra running beat merely renders late. A terminal beat settles the row directly under the lane proof its binding gives it (not the main lane a notification would use, and not by card id, since the engine is naming a child rather than closing a card), freezes the inline group-row twin in the same beat so a later sweep cannot reset the real tool count, clears the session-resident ledger, and fires the stop hook only for a child whose start really fired. It does not mark the row live or emit a second progress beat, and it shares the settled-row ledger with the other two settle legs so a replay or a double-delivery cannot produce a second end. Three things are pinned **unchanged**: a running beat, an absent status (older engines never send the field, and reading absence as terminal would make every child row disappear on its first beat), and the workflow lane gate, which still runs before any of this |
 | `scripts/run-assistant-arm-identity-test.mjs` | The identity keys on an assistant row, and an explicit account of the two that are **deliberately not** there. What the renderer received was a bare role-and-content object, so a dozen consumer sites downstream were each estimating what the message envelope should have told them. The id is taken from the engine's own event id rather than minted locally, because it has to be **the same value** on the live leg and on a durable replay — a freshly minted one would make a replayed message look new to a host's dedup and to rewind — and when the wire carries none the key is simply absent rather than filled with a random stand-in wearing an identity it does not have; it is also kept distinct from the envelope's own local render key, which is a different identity. The model name comes from what the host pinned when it opened the stream (the request was the host's to build) and is never guessed, since a wrong model name is worse than none once a billing or capability face looks it up. Usage and stop reason are **not** minted on this arm, and the reason is frame order rather than effort: content arms arrive before the turn's closing frame, so at the moment the arm is emitted the engine has not yet said what the round cost — anything put there would be an estimate, which is the very thing this work exists to remove — and synthesising a follow-up assistant update when the real figure lands is also refused, because that shape does not exist upstream and would place a message in the transcript the engine never sent. Their real values leave through the turn's own neutral arm as two superset keys, the usage one reusing the **same single mint point** the footer rollup already folds so the two faces cannot diverge, and the stop reason passed through verbatim as an open set — the machine signal for *was this turn cut short*, previously blind on both the stream and the trace. The existing behaviours beside them are pinned too: no arm at all when usage is wholly absent, and the sub-flow cut-out that keeps a child's turn from driving the leader's face |
 | `scripts/run-text-segment-authority-test.mjs` | The **authoritative segment replacement** on `text_end` (server >=7.75.3). `text_end.content` now goes through the same redactor as `result` and the ledger while `text_delta` stays verbatim, so the two **may differ** — an answer that quoted a credential used to be committed to the local transcript in its unredacted form, because the arm only forwarded the boundary signal. Six timing shapes are pinned, two of which an adversarial review reproduced against the installed engine's real bytes and which the first design got wrong in both directions: a second boundary in the same turn (the per-block case on one provider lane) used to make the first segment's prose vanish, and a boundary that arrives *after* the tool card (the other lane emits it at finalize) used to be read as "this package never handled that segment" and reported nothing at all. Three additive keys, all never-false; the two shapes that look alike are told apart by the second one, because the host's action in them is the opposite. The end-to-end legs drive the real pipeline without hand-inserting a segment commit — doing so is exactly what hid the first defect. A second review round then found two combination timings on top of the first fix — a tool card followed by *more* deltas in the same segment, and a byte count that had been documented as a message count — and both are pinned here too. A third round caught a length that the prose called bytes while the code returned UTF-16 units — harmless in ASCII, and on CJK text enough to leave the credential on screen — plus a backfill ledger that had to be kept in step, so the terminal frame does not re-render the segment a second time — kept in step only where the whole stretch sits in one message, because those ledgers are per-message and a fourth round showed that writing across them charges one message's prose to another. A fifth round settled the whole class into one invariant the guard now checks against the previous release's behaviour: this package only rewrites bytes it is still holding in the current message — once a segment has crossed a package-side boundary it emits the three keys and changes nothing else **0.68.2 (CC-01):** the segment identity is now minted here, not by the host: every committed assistant text row carries a top-level `_sema_segment_id` (stamped once at the `adapt()` exit, so the durable whole-message leg and the streamed-segment leg are covered alike; thinking blocks, tool_use-tailed rows and chrome events are left byte-for-byte), `text_segment_end` carries the same value as `segmentId` before rotating, subagent boundaries never rotate, and a replayed stream yields the same identities. Three mutations (no rotation / no stamping / stamping tool_use rows) each turn the guard red **0.69.0 (CC-02):** the same authority replacement now covers the reasoning face (`reasoning_end`, server >=7.77.0): a thinking block still buffered is swapped whole and its live tail recomputed; one already committed at a boundary (the usual timing, since the first text delta commits it) is left untouched and the host is told the row to replace by its uuid, never re-emitted. Subagent boundaries are ignored and the text-segment identity does not rotate **0.69.1 (CC-09):** the run-stream replay guard still drops a frame whose event id was already seen, but it now reports the drop through the host's dropped-frame sink as `duplicate_seq` instead of vanishing silently (server 7.77.0 reuses the first reasoning delta's id for `reasoning_end`, so that authoritative segment is lost on the print lane until 7.78.1); the interactive adapter has no such guard and keeps receiving it **0.69.1 (CC-10/CC-11):** subagent segment-end frames are fenced on all three identity keys (a frame carrying only `sourceTaskId` no longer masquerades as the leader's), and a reasoning segment that spans tool cards now hands the host every committed row it covers (`committedUuids`) so nothing unredacted is left behind |
 | `scripts/run-gate-negative-controls-test.mjs` | Whether the registry-shaped guards among the suites above actually turn red when the material they check really breaks — a census found the ones clean enough to rehearse safely (closed sets, mirrors, baselines, floors, a type-shape ratchet) without touching any judgement code. Each is exercised by tampering a disk copy of the real material, spawning the guard's own unmodified script, asserting it exits non-zero and names the disease, then restoring the file byte-for-byte. The guards of the same shape that are too costly to rehearse, and the behaviour/projection suites that red on their own assertions, are catalogued rather than rehearsed — see `docs/GATE-NEGATIVE-CONTROLS.md` for the three tables (all generated from `gates-manifest.json`, which carries each suite's classification), the reasons, and a one-minute manual replay recipe for each blind one. The suite reconciles its case list against that classification by name in both directions, so a case quietly dropped from the array without the manifest following is itself an undeclared blind guard. The backup that makes the restore possible is taken by **exclusive create**: checking for it and then copying are otherwise two steps, and two instances can pass the check together — the later one overwrites the only clean copy with material the earlier one has already tampered, and the rehearsal that promises to leave no trace leaves a permanently corrupted file instead. That interleaving is rehearsed too, in a throwaway directory of its own. A run that is interrupted — by the suite runner's time limit, by Ctrl-C or by its terminal closing — restores what it was in the middle of tampering with before it exits: the rehearsed guard's whole process group is stopped first, so nothing that guard started can write the tampered state back afterwards; the file is restored byte-for-byte from its backup and the backup removed only once the bytes match; the build output is regenerated from the restored sources within the grace period the runner allows (a rebuild that would overrun it is cut short, its compiler stopped, and the run says the build output was not regenerated); and the run exits with 128 plus the signal number rather than passing. Before the first rehearsal the suite also looks for backups left behind by an earlier run that could not clean up — in the trees its targets live in, following a symlinked dependency directory — and if it finds one it refuses to start, names the file and how to check and restore it, and leaves the backup untouched rather than overwriting what may be the only clean copy. |
-| `scripts/run-engine-cap-reader-factory-test.mjs` | The one shared implementation behind every capability reader's four ports (cache, generation gate, probe tee, invalidation), exercised as a table: every reader in the table runs the *same* criteria (the table length is the source of truth, and a roster check fails the gate if any source file calls the factory without having a row) — the four states, a throwing projection treated exactly like an unreadable one (and never escaping the tee), the generation rules (a stale generation is dropped before the projection even runs; a projection that changes the generation mid-flight cannot overwrite the newer value, whether it returns or throws; the caller's `opts` is snapshotted once; and omitting the generation still writes, because that supply is additive — except that a write whose projection itself triggers an invalidation is dropped either way), the top-level key's getter being read exactly once, a freshly minted "unobserved" reading on every miss (two misses are never the same object, so a consumer that mutates one cannot taint another base URL), one independent table per reader that never take each other down, invalidating one base URL leaving every other base URL's reading untouched, `forget`/tee being no-ops on an empty or non-string base URL, the read anchor being resolved dynamically, and the deliberate split in how presence is judged per reader. A stored reading is frozen at every level, whether it reports a value or reports that the position was not reported, so a consumer that writes to the object a reader hands out changes nothing for later readers (a strict-mode write throws); an unobserved reading is minted fresh on every read, and the capabilities body passed in is never frozen. Every stored reading also carries the identity of the capabilities-cache generation it was written under: once `invalidateEngineCaps` advances that generation, every reader answers "unobserved" without any `forget*` call, until the current generation writes again — checked against a late response from the previous engine (dropped when it carries the generation taken when its probe started; written as current when it carries none), a replacement probe that succeeds, throws, returns a bad shape, hangs or never comes, two readers interleaving on one base URL, two base URLs written in either order, two in-flight probes landing in either order, and a projection that invalidates midway — and a test reset of the cache never revives an older reading; a host that never invalidates is checked step by step against the previous implementation on a seeded random sequence, and the downstream decisions that depend on these readings are pinned at each moment around a switch. |
+| `scripts/run-engine-cap-reader-factory-test.mjs` | The one shared implementation behind every capability reader's ports, exercised as a table. Since 0.95.0 a reading is a lazily computed, frozen projection of the capabilities cache's landed response for that base URL, memoised per landing; the probe tee port is a retired no-op and the per-reader `forget` port is gone — invalidation is `invalidateEngineCaps` alone. Every reader in the table runs the *same* criteria (the table length is the source of truth, and a roster check fails the gate if any source file calls the factory without having a row): the four states (nothing landed, present, key absent so not reported, malformed so unobserved); a throwing projection or descriptor read treated exactly like an unreadable one and never escaping; a projection that changes the generation mid-flight answering unobserved without being remembered; a getter at the top level or at any nested position never being executed and reading as malformed; a freshly minted "unobserved" reading on every miss (two misses are never the same object, so a consumer that mutates one cannot taint another base URL); one independent memo per reader that never takes another down; invalidating one base URL leaving every other base URL's reading untouched; `invalidateEngineCaps` and the retired tee port being no-ops on an empty or non-string base URL; the retired per-reader `forget*Reading` exports being absent from the root entry; the read anchor being resolved dynamically; and the deliberate split in how presence is judged per reader (own data property only, so a key that exists only on the prototype is not reported). A stored reading is frozen at every level, whether it reports a value or reports that the position was not reported, so a consumer that writes to the object a reader hands out changes nothing for later readers (a strict-mode write throws); the capabilities body passed in is never frozen. A reading follows landings and invalidations rather than any per-reader state: after `invalidateEngineCaps` every reader answers "unobserved" until the next landing, checked against a late response from the previous engine (dropped by the cache), a replacement probe that succeeds, throws, returns a bad shape, hangs or never comes, two readers sharing one landing, two base URLs landing in either order, two in-flight probes landing in either order, and a projection that invalidates midway — and a test reset of the cache never revives an older reading. A host that never invalidates is checked on a seeded random sequence (each read is the projection over the latest landing, one object per landing, getters never executed), and the downstream decisions that depend on these readings are pinned at each moment around a switch. |
 | `scripts/run-mcp-liveness-test.mjs` | The engine's **liveness observation** about each MCP server it hosts (`wiring_manifest.mcp[].liveness`, engine-side from core 7.24.3 / server 7.91.2): one reader, one word list, one leg-level verdict. The cell answers *can this server still be reached* — it is not the connect-time verdict beside it, which the engine deliberately freezes (a server that died mid-run still reads `connected`), and it is not a re-dial's judgement either, so `status: "failed"` next to `liveness.state: "reachable"` is a **real row**: the server answered the handshake and answered with a protocol error — up, and misconfigured. The three words are read as a closed set (an exchange completed / it was lost in transport or the clock / what came back does not answer the question), and a word outside it is malformed rather than rendered, because a word nobody upstream has defined is not a sentence worth putting on screen. **Absence is the fourth reading and is not one of the words**: it means *no liveness record is available*, which on the wire covers a server this leg never reached, a declaration that could not be dialled, an older engine, and a record the projection ahead of us dropped — all indistinguishable, so it is never read as "we looked and could not tell" (a strictly stronger claim), never as healthy and never as off. The failure-class footnote rides the unreachable word only, and one that turns up anywhere else, or that is malformed, loses **just the footnote** while the word and its timestamp stay: the honest reading is then "cannot be reached, reason not given", not "this record is broken". Malformed never becomes healthy: a cell that is present but unreadable marks its row and pulls the leg-level verdict back to *cannot tell*, since letting it sit beside a reachable row would report a leg as reachable on the strength of a record that may well have said the opposite. The verdict takes the worst fact first rather than a majority or the newest reading, carries no server count — so there is no fabricated zero to be read as "no problems" — and its timestamp belongs to **that leg's** observation, not to now: the engine runs no probe and adds no traffic of its own, so this is a per-leg snapshot rather than a heartbeat. The replayed roster on the session panel and the live leg go through the same reader, and the panel's own deployment-side rows carry no liveness position today, so the verdict does not borrow a word from that face. The word list is bitten in both directions where an installed witness exists and the absence of one is itself asserted against the installed engine's version, so the day it ships the comparison starts on its own; since sdk 11.2.0 the wire types declare the cell, and the word list and the presence test are pinned to those types at compile time in both directions, so the two can never drift apart |
 | `scripts/run-peer-lane-rules-write-capability-test.mjs` | Two more engine self-descriptions read the same four-state way as their seven sibling capability readers (`capabilities.peerLane`, `capabilities.permissionRulesWrite`): an absent key is not reported (an older engine that predates the position, never folded into `false`), `true` is present, `false` is a positive absent (the cross-session lane not being mounted on this deployment, or this particular call not being able to reach the tightening-direction write entry), and anything else is unreadable and drops the cell. Each carries its own single-source verdict (`peerLaneAvailable` returns `yes`/`no`/`unknown`; `permissionRulesWriteAvailable` collapses to a plain boolean, present being the only `true`). The write-entry position pairs with a boolean convenience port in the persisted-rules module, and this guard pins that port to derive from nothing but this one reader's own reading — never a conjunction with the lane-reachable position, and never a second read of the deployment-level existence signal the revoke surface uses (the two are documented as reading differently on purpose): a deployment where the lane answers true but the write entry's key is simply absent (an older binary) must still come back `false`, a deployment where the write entry answers true while the lane key is entirely unseen must still come back `true` (proving no silent conjunction crept in), seeding only the general capabilities cache — never this reader's own feed — must still come back `false` (proving the convenience port cannot be satisfied by the wrong table), and passing an explicit `undefined` base URL must still come back `false` even while a different, already-installed engine target answers `true` for the same position (an adversarial pass found the naive forward of that parameter falls through to the reader's own convenience default, silently answering for whichever engine happens to be installed rather than the caller's absent target — the fix routes an explicit absence through the same empty-string path the reader treats as unobserved). |
 | `scripts/run-lane-proof-identity-test.mjs` | The **instance identity of a lane proof**: the main-lane proof is minted fresh on every emission. Previously a single module-level constant object was handed both to `laneOf(an unregistered task id)` and to some fifty main-lane emission points, so two unrelated consumers — across adapter instances, across streams, across turns — held the same object: writing a card id onto one of them was readable on the other, and the four opening main-lane events changed together. Nothing in this package writes to a lane proof and the known consumers only read it, so this is an **aliasing hazard on a published output surface** rather than an observed corruption — a consumer that uses the proof as an identity key, for dedup, or as a view-layer identity would conflate two unrelated rows without writing a single byte, which is precisely the half that freezing the object would not solve. The gate therefore anchors on instance identity: two independent adapter instances, two rows inside one instance, the same id read twice, and two arms in one beat are each distinct references; mutating one leaves the others byte-identical; and the subagent lane, which already minted fresh, is the control that proves the criterion discriminates. The main-lane **value** is unchanged — an unregistered id still answers `{lane:"main"}` with exactly one own key and still emits its events, so absence is not turned into a second kind of absence — with ordering pinned three ways (registered-then-read, read-then-registered with no retroactive edit of an already delivered proof, the same id twice) and the id failure classes pinned four ways (unregistered, empty string, absent, non-string, the last two emitting no panel event at all rather than an ownerless proof). Where one row emits **two** events — the terminal-tick and card-close legs, which each yield a lifecycle stop and a panel end — the attribution is decided once (a consumer binding a card between the two yields must not split one row across two lanes) while each event still gets its own proof, so a host consuming them one at a time cannot poison the second before it is even yielded. The run stream leg is covered as the same shape, and a syntax-tree check forbids reintroducing a module-level lane-proof object literal or a module-level `LaneProof`-annotated binding (judged on the type node, not on text, so a compile-time pin tuple that merely mentions the type is not miscaught), backed by a type-checker pass that also catches an un-annotated module-level cache such as `const x = mainLane()` while letting the callable factory itself through, while the module-private three-state sentinels of the untrusted read are frozen instead — only `Object.freeze` counts, never `Object.seal`, which still permits writes to existing keys — their exposure being confined to one module |
 | `scripts/run-memory-entries-wire-test.mjs` | The two memory-governance capability bits and the three memory-entry response readers. Each bit (`capabilities.memoryCompliance`, for the entry-provenance and erasure endpoints; `capabilities.memoryOrigin`, for the external-origin listing and clearance endpoints) is read the same four-state way as its sibling capability readers: an absent key is reported as not reported (never folded into `false` — an older engine simply does not answer, and the right next step is to try the endpoint and read its 501), `true` is the face being mounted, `false` is a positive "not on this deployment" (the wire does not distinguish a backend without control-plane ownership from an empty operator roster, so the wording never guesses which), any non-boolean value is unreadable and drops the cell instead of being folded into "absent", and a capabilities body that is not an object at all is unreadable rather than "not reported". The two bits deliberately stay **two** readers with two separate per-engine tables, because the engine deliberately keeps them two separate product faces even while they happen to carry the same value today: feeding one an unreadable body, or invalidating one, leaves the other's reading untouched, and a body where one is on and the other off is answered one bit at a time. The entry-export reader narrows each row on its own (an empty array really is zero rows, a non-empty array with nothing readable in it is reported as unreadable rather than as "no rows", and partly bad rows are kept with a dropped count), reads the external-origin marker as three states rather than a boolean (the two structural carriers mark a row; a row whose frontmatter cannot be read, or which carries the third, suspended-form carrier, is undecidable, because the judge for that carrier lives in the engine and this package refuses to mint a second copy of it), and treats an unreadable "is this the whole scope" flag as "not the whole scope". Its verdict port implements — in code, not in a comment — the rule that an empty answer is never a clean store: the caller must state whether the request declared origin-awareness, because this endpoint withholds marked entries by default and the two bodies are shaped identically, so without that statement an empty answer is only ever "unknown"; the affirmative answer is scoped to the one named scope and carries that scope with it, and the type has no store-wide arm at all. The erasure receipt reader keeps three things apart that are easy to collapse: "this call erased nothing" (a real receipt whose erased list is empty and whose not-found list explains why, per id), "a 200 with an empty body", and "a body that could not be read" — at the reading, the counting and the verdict layer alike; it refuses a version envelope it does not recognise instead of reinterpreting it, treats the three closed vocabularies as closed (an unknown word is unreadable, never folded into a known arm), keeps an unreadable binding as unknown instead of claiming "unbound", passes the "history cannot be judged" flag through as four states (set, explicitly unset, absent, and present-but-unreadable — an unreadable flag is kept distinct from an absent one, and the history verdict then answers "unknown" rather than the stronger claim), and answers the replay question as three states so that the degraded lane is never retried automatically. The clearance receipt reader carries the cleared marker through verbatim and says separately whether it was reported at all. Every array in every response is snapshotted once — the length is read exactly once and each index exactly once, rather than iterating the caller's own iterator — because an array that reports one length while being walked and another afterwards could otherwise have a marked row quietly dropped while the "was anything unreadable" check saw nothing, which ends in calling the scope clean; an array that reports an absurd length is reported as unreadable rather than silently truncated to its first rows. All three readers never throw. |
-| `scripts/run-memory-verbs-wire-test.mjs` | The five memory-governance verbs as call ports — entry provenance, compliance erasure, the external-origin listing, the clearance ledger and the un-mark valve — on top of the readers above. One failure judge serves all five, and its first question is **provenance, not status**: the engine stamps a machine code on every refusal it mints, so a 501, 405, 409, 404 or 400 that carries **no code** proves nothing about who answered — a proxy or gateway returning the same status may well have passed the request on first — and every such answer is reported as "no verdict" rather than as "nothing happened". Twenty-one coded refusals each get their own arm, branched on the code alone: the status cannot tell them apart (nine different operator actions ride the same 409 here), and conjoining the status would silently demote a refusal the day the engine moved it. A coded 5xx, a coded answer with no status at all, and a coded 4xx this version does not recognise all land in the "cannot tell" arm, because on a non-idempotent verb the default for "could not classify" must be "do not know", never "did not happen". The judge is called from catch blocks, so each of its own property reads is guarded: an error object whose accessors throw is classified, not re-thrown. The capability gate runs before the call and reads the two bits the engine keeps deliberately separate (one for provenance and erasure, one for the origin faces); only an engine that positively says the face is off stops the request, while "this binary does not report that bit" and "this process never saw a capabilities body" both still send — folding "cannot say" into "is not there" is the dishonest-absence shape this package refuses, and these routes answer the capability gate before touching anything. A missing capability reading is a named, explainable error rather than a silent default that would answer for whichever engine happens to be installed. Each verb returns its own discriminated union whose success arm, refusal arm and cannot-tell arm share no keys, so a consumer cannot express "could not read it" as "it worked". The two write legs reuse the erasure and clearance receipt readers rather than minting a second copy, which keeps "this call erased nothing", "a 200 with an empty body" and "a body that could not be read" three separate things here too; the provenance account is narrowed only to its envelope and discriminants and otherwise passes through verbatim, and an account stamped with a newer envelope version is refused rather than reinterpreted. All three receipt-bearing verbs additionally reconcile identity — the account id, the attestation request id and the clearance receipt entry id must be the ones that were sent — because a readable receipt is not yet a receipt about this call. The origin listing carries the server's own echo of the scopes it actually audited, the type has no store-wide arm, and the coverage port takes a mandatory second argument and has no "clean" arm at all: the strongest thing it will say is which scopes were audited. Neither write leg is ever retried, including the refusal whose documented recovery is to send again, because that resend completes whichever clearance row is already open and the audit attribution on it is a person's signature. Request bodies are handed over verbatim — the degraded-erasure authorization is never injected — while the scope list is sent as the snapshot this port validated, so an array that reports one length while being read and another afterwards cannot make "the list I checked" and "the list I sent" two different things. An empty operator list (501 `capability.operator_principals_required`) is a refusal before anything ran (`operators_unconfigured`), not an unknown outcome. |
+| `scripts/run-memory-verbs-wire-test.mjs` | The five memory-governance verbs as call ports — entry provenance, compliance erasure, the external-origin listing, the clearance ledger and the un-mark valve — on top of the readers above. One failure judge serves all five, and its first question is **provenance, not status**: the engine stamps a machine code on every refusal it mints, so a 501, 405, 409, 404 or 400 that carries **no code** proves nothing about who answered — a proxy or gateway returning the same status may well have passed the request on first — and every such answer is reported as "no verdict" rather than as "nothing happened". Twenty-one coded refusals each get their own arm, branched on the code alone: the status cannot tell them apart (ten different operator actions ride the same 409 here), and conjoining the status would silently demote a refusal the day the engine moved it. A coded 5xx, a coded answer with no status at all, and a coded 4xx this version does not recognise all land in the "cannot tell" arm, because on a non-idempotent verb the default for "could not classify" must be "do not know", never "did not happen". The judge is called from catch blocks, so each of its own property reads is guarded: an error object whose accessors throw is classified, not re-thrown. The capability gate runs before the call and reads the two bits the engine keeps deliberately separate (one for provenance and erasure, one for the origin faces); only an engine that positively says the face is off stops the request, while "this binary does not report that bit" and "this process never saw a capabilities body" both still send — folding "cannot say" into "is not there" is the dishonest-absence shape this package refuses, and these routes answer the capability gate before touching anything. A missing capability reading is a named, explainable error rather than a silent default that would answer for whichever engine happens to be installed. Each verb returns its own discriminated union whose success arm, refusal arm and cannot-tell arm share no keys, so a consumer cannot express "could not read it" as "it worked". The two write legs reuse the erasure and clearance receipt readers rather than minting a second copy, which keeps "this call erased nothing", "a 200 with an empty body" and "a body that could not be read" three separate things here too; the provenance account is narrowed only to its envelope and discriminants and otherwise passes through verbatim, and an account stamped with a newer envelope version is refused rather than reinterpreted. All three receipt-bearing verbs additionally reconcile identity — the account id, the attestation request id and the clearance receipt entry id must be the ones that were sent — because a readable receipt is not yet a receipt about this call. The origin listing carries the server's own echo of the scopes it actually audited, the type has no store-wide arm, and the coverage port takes a mandatory second argument and has no "clean" arm at all: the strongest thing it will say is which scopes were audited. Neither write leg is ever retried, including the refusal whose documented recovery is to send again, because that resend completes whichever clearance row is already open and the audit attribution on it is a person's signature. Request bodies are handed over verbatim — the degraded-erasure authorization is never injected — while the scope list is sent as the snapshot this port validated, so an array that reports one length while being read and another afterwards cannot make "the list I checked" and "the list I sent" two different things. An empty operator list (501 `capability.operator_principals_required`) is a refusal before anything ran (`operators_unconfigured`), not an unknown outcome. The store-form refusal family (a directory in the store that leads out of it, a scope home that is no longer a real directory) is read as one `cannot tell` arm, because the engine gives no marker for whether the refusal came before or after the commit point. |
 | `scripts/run-dist-orphan-test.mjs` | Every `.js` / `.d.ts` under `dist/` must have a same-named source under `src/`, and every source must have its build output — because the compiler only writes and never deletes, so a module removed from the sources keeps shipping from the previous build (the whole `dist/` directory is on the publish whitelist) while the public-surface gate only looks at what the barrel exports and the hygiene gate only looks at forbidden words. Orphans are named one by one; the pre-publish posture is a clean rebuild, and this gate is the check that the posture was actually followed. |
-| `scripts/run-dist-comments-test.mjs` | **dist ships zero comments.** Since 0.77.2 the build strips comments (`removeComments`); this gate walks every shipped `dist/**/*.js` / `*.d.ts` and counts comment trivia with the TypeScript scanner (string literals containing `//` and generator methods are not comments), failing on the first one (`DIST-COMMENT-FAIL`). Source comments are an internal surface; what still ships is code, string literals and type-level text, which the hygiene gate screens. Negative control: one plain comment appended to `dist/index.js` turns it red. |
+| `scripts/run-dist-comments-test.mjs` | **dist ships zero comments.** Since 0.77.2 the build strips comments (`removeComments`); this gate walks every shipped `dist/**/*.js` / `*.d.ts` and counts comments with the TypeScript parser (a comment is whatever sits between two tokens, so string, template and regular-expression literals that merely look like comments, and generator methods, are not comments), failing on the first one (`DIST-COMMENT-FAIL`). Source comments are an internal surface; what still ships is code, string literals and type-level text, which the hygiene gate screens. Negative control: one plain comment appended to `dist/index.js` turns it red. |
 | `scripts/run-task-request-omission-receipt-test.mjs` | Where every key a client hands to the request constructor ends up. The constructor used to answer "not stamped" the same way for four different reasons — value absent, no such row, wrong lane, live gate closed — and a key it had never heard of did not even get that: an unattended run could pass a system prompt, an output schema and a spend cap and receive a body holding the objective and the session id, with nothing anywhere saying what was left out or why. The guard pins the three answers apart. **Seated** keys reach the body verbatim on the unattended lane. Keys the package **knows but did not carry** never throw, never reach the body, and each gets a receipt row with one word from a frozen cause list — every present key is on the body or on the receipt, never both and never neither, checked across both lanes with the live gate open and closed against a key-by-key table written independently of the package's own routing. Keys the package **does not know** are refused loudly and are a separate cell, not a fourth cause: the cause list has no word that could hold them, and the seat reader answers `unknown`, not `none`. The cause list is bitten from both sides (exact, every word producible, nothing produced outside it, the judge table's keys read from source through the TypeScript parser) and no second hand-copied list may exist in `src/`. Upstream claims are read straight off the installed SDK typings: a key seated in this release must be a named request field, a key registered as having no upstream counterpart must not be — the day it appears the guard turns red — and the index signature counts as evidence for nothing. The per-run file-history opt-out word is covered the same way: seated on the two user lanes behind the live gate and absent from the side-channel lane, stamped only for its one legal word, refused loudly for any other value, and read after every existing declaration so that it cannot erase them. The wording shown to end users is pinned alongside the wording for integrators: a second sentence table, keyed by the same frozen cause list and exhaustive over it at compile time (the check sits on a two-column table literal — a lead-in plus a reason clause — from which the full sentences are derived), gives each cause one plain sentence that says the item was not sent and why (the integrator table now carries the same literal-level check — before, a missing sentence failed the build but an extra one did not), with no integrator vocabulary and no claim about whether anything took effect; every sentence differs from the others and from the integrator wording and passes the published-text hygiene list, and an unknown word, a non-string, a throwing object or a prototype-chain name gets one honest generic sentence instead of a throw, an echo or an invented cause. The reason clause on its own is available for clients that already print their own "was not sent" line: the full sentences and the generic fallback stay byte-identical to the previous release, every clause starts lower-case with no closing full stop, an unreadable cause gets the generic clause, and no full sentence may appear as a literal anywhere in the source. When a consuming client's own copy of the cause table is on disk, the guard also checks that every word in it is one of the package's causes, with no word listed twice (a word only the client's table has fails; a word only the package has is reported, not judged), and reports that check as skipped rather than passed when the table is not there. A request key the engine retired and refuses by name is, from 0.92.0, refused at construction by the same unknown-key rule as any other key the package does not know (it names the key and never echoes the value), on every lane, in every live state and through both doors; the seat reader answers unknown and the wire-side judge names it. |
 | `scripts/run-session-policy-wire-test.mjs` | The per-session tool-rule face: the capability bit that says whether an engine keeps such rules at all, and the narrow read plus tightening orchestration built on it. The bit is read the same four-state way as its sibling capability readers — an absent key is not reported (this binary predates the position itself, which says nothing about whether the face exists), `true` is present, `false` is a positive absent (this deployment keeps no per-session rules), any non-boolean value is unreadable and drops the cell rather than being folded into "absent", and a capabilities body that is not an object at all (an array included) is unreadable rather than "not reported". Its single-source verdict answers whether to show the tightening entry: only an engine that says yes is `yes`, both a positive no and a binary too old to answer are `no`, and never having observed a capabilities body is `unknown`. Whether to put a request on the wire is deliberately a **different** question with a different answer for that last state, and lives with the orchestration. The read narrows three ways that must not collapse into each other: a record that really is empty (present, version zero — what an engine answers for a session no rules were ever written for), a record that cannot be read, and a call that failed with a typed disposition. A half-bad record — one rule bucket well-formed and another the wrong shape — counts as unreadable in full, because the write verb replaces the whole record: dropping the bad bucket and writing the rest back would empty it, which relaxes the rules while the caller sees a 200. An unreadable version stamp is never filled in with a zero, a bucket that reports an implausible number of entries is unreadable rather than walked or truncated, each array's length and each of its indices are read exactly once, and a throwing accessor is unreadable rather than propagated. Every load-bearing key is read as an own property — the envelope, the version stamp, each of the five buckets and each array index — because a prototype lookup would let a polluted prototype put a bucket into the reading that the wire never carried, and since the write replaces the whole record the union would then write that invented restriction back as a real one; a guard pollutes the object and array prototypes in place and proves all four shapes stay out. Because the write replaces the whole record, adding a restriction means writing "what is already there, plus the new entries": the union only ever adds, de-duplicates verbatim, keeps a bucket that is present but empty (present-and-empty and absent are opposite meanings, and dropping it would relax the rules), mints no bucket neither side had, copies entry bytes as they came (no trimming, sorting or path rewriting — those judgements belong to the engine), and takes its bucket names from the engine's own type surface rather than a hand-copied list, so a new bucket upstream is a compile error instead of a silently dropped one. Which differences count as relaxing is the engine's judgement and is never re-implemented here: a refusal on those grounds is reported verbatim, never swallowed and never retried. The orchestration is guarded on three axes. Timing: when the record moves between the read and the write, it re-reads and re-writes **exactly once** — two reads and two writes, no more — and the second attempt's union carries the other writer's entries, which is the entire point of re-reading; a second collision is reported rather than retried a third time, and an uncontended write makes exactly one round trip. Concurrency: an explicit barrier holds both orchestrations first reads at the same version before either may write, and the criterion is how many times the store actually rejected a stale version rather than how many writes it saw — the latter is equally true of two serial successes, so it would stop detecting contention the day the interleaving changed. Under real contention the store rejects exactly once, both writers land on strictly different versions, both writers entries survive in the final record, and the round trips are exactly three reads and three writes; the same two orchestrations run serially are asserted to reject zero times in two reads and two writes, which is what proves those numbers are discriminating. On a store where every write loses the race both report a collision having written exactly twice each. Failure classification: a relaxation refusal, a refusal to stamp a version the store cannot establish (the same status code as the relaxation refusal but a different machine code, and folding it into that arm would send the caller off to edit entries that are not the problem), a missing session, a deployment without the face, an ownerless session, a collision code, a bare conflict with no machine code, a rejected body and an unauthorized call each land on their own arm — the collision arm is matched on the machine code verbatim rather than on the status, because two different situations share that status and only one of them is worth retrying. The remaining split is not "which code is this" but "did the engine answer at all": an answered client-side refusal is allowed to say nothing was written, because every such refusal on this endpoint is emitted before the record is touched, while a throw with no answer at all — a dropped connection, a timeout, a response body the transport itself could not decode, a server fault — can only say "unknown", since that throw may well have happened after the record was already saved. Two guards prove that is not theoretical: a write whose receipt cannot be read, and a write that throws after the fixture store has committed, both leave the record changed. Neither is success nor failure: the only honest answer is "unknown", it carries no version, and it is never retried. The direction of the change is likewise never claimed. The engine’s tighten-only rule is an identity gate, not a field gate — for a principal the deployment treats as an operator it does not run at all, so a union that adds a name to an existing allowlist is accepted and really does widen it. This package does not mint a second copy of that rule, so what it reports is the fact it can stand behind: the record now holds what it already had plus the entries sent here. The sentence for a saved write is pinned to contain no claim of tightening, narrowing or restriction, and a guard reproduces the operator case to prove the widening is real while the wording stays honest. Every sentence the module mints is checked pairwise distinct, with the receipt-unreadable one required to keep its "may already be in effect" and the record-unreadable one required to say nothing was written. When a read comes back without a usable record, one sentence per failure kind and per unreadable position says so in plain words, never naming internal codes, never saying the record is empty, and falling back to one general sentence for anything it does not recognise. A sixteenth sentence covers a session that has not begun on the engine yet — a state only the host knows, since the server gives the same answer for a session that does not exist and for someone else's session — and it is checked like the others: distinct from the other fifteen and from the general fallback, free of machine words, never claiming the record is empty or that nothing is restricted, never saying the contents are unknown (there is nothing to know yet), read from own data properties only, and compared with the previous release's package so that the other fifteen sentences and the general fallback stay unchanged byte for byte. |
 | `scripts/run-persisted-rule-write-test.mjs` | The **single-step tightening write** for persisted permission rules — the dual of the revoke surface, and the half where a hopeful reading is expensive. The two behaviours this entry accepts are **derived** from the three-state vocabulary by subtracting the widening one, never hand-copied: the guard bites in both directions (every word in the derived table is really accepted; every constructed outsider — casing variants, trailing whitespace, the widening word itself — is refused before a single round trip), keeps a word-count canary against the parent table, and pins that the source file contains **exactly one** array literal carrying two or more behaviour words, so a second hand-written table shows up as a boundary failure rather than as drift nobody reads. A standing approval is minted by answering a permission question or by importing settings; this entry is not a third route, and the widening word is unspellable in the type. The outcome is a discriminated union whose two failure arms are **not** interchangeable: ten refusal causes each promise the same single thing — not one byte reached the store — while three separate words say the opposite, that the outcome could not be read at all. The service's own "I cannot tell" (a write that could not be confirmed as standing: store wobble, a redemption leg with no decidable ending, or a write that landed and was revoked concurrently before the read-back) stays in the second group, because announcing "nothing was written" invites a clean retry that is not clean, and announcing success misreports a tightening that may already be gone. Anything the shared failure classifier does not recognise defaults to the same place — this is a non-idempotent verb, so "unclassified" must mean "unknown", never "no write": a 500 can happen after the store commits. A 2xx whose body cannot be read is pinned in the same direction and from both sides: it reads as unknown, and the unknown arm carries **neither** the revision nor the written row, so a consumer cannot even spell the shape that would let "unreadable" pass for "written". `persisted` is guarded against the reading everyone reaches for first: it says *this call wrote*, not *a new rule now exists* — an equivalent rule already in the store still mints a fresh causal point, so the lane honestly reports `persisted`, and the material for judging whether the **logical** rule is new (the approval ledger on the returned row) is handed to the caller rather than folded into the discriminant, since the package does not have the one fact that judgement needs. The returned row goes through the **same single narrower** the listing surface uses — proven by running one row corpus through both legs and asserting the two verdicts agree entry for entry (an adversarial pass that forks the listing leg back into an inline copy reds here immediately), plus a source pin that the predicate is defined once and called from exactly the two legs. That sharing is what keeps a row whose behaviour cell is unreadable **visible in both places** rather than hidden by one of them — and the shared narrower is deliberately followed by a second, *different* question only the write leg can ask: is the row that came back **the rule that was just sent**? A rule's identity is a triple, so a receipt missing its behaviour cell, carrying the sibling state, carrying the widening one, or naming another text or another scope is not evidence that the requested tightening is standing; it reads as unknown with its own word, kept distinct from "unreadable" so the two stay tellable apart, while display and derived cells may vary freely. An adversarial pass found both of the gaps this pins: the receipt check that only looked at whether the row was renderable, and a subtler one — pulling the verb off the injected port and calling it bare drops the receiver, so a host that hands over a real resource object (a class instance whose verbs reach the transport through `this`) would see every write throw and be reported as "could not tell", retry after retry, while the package's three other ports call their verbs as methods and work fine. Both are pinned from the failing side: a shorthand-method facade and a class-instance facade must reach the transport and return a real outcome, with the bare-call throw proven to be a real failure mode first. A 405 is split in two, because only the engine's own bare code is evidence about **the engine**: with it, the path exists and this verb does not, so this worker predates the verb; without it — an absent, empty, or foreign code, which is what a proxy or gateway blocking the method typically returns as HTML or an empty body — what was seen is that the verb was refused, while **who** refused it and **at which hop** is unknown, so it lands in the unreadable-outcome arm with its own word rather than sending someone to upgrade a worker that is fine, hiding an entry that is live, or — the part a second adversarial pass insisted on — promising that nothing was written. That promise is what the refusal group means, and a middlebox is free to forward the request and only then answer 405 on its own policy, so a caller who skipped reconciliation on that word would leave behind a standing refusal the user believes never took effect; the guard pins exactly that shape, with a double that writes the rule and *then* answers 405, and with the engine's own bare code still landing in the refusal group beside it. (The same passes caught the naive status-only reading and the asymmetry where an empty-string code fell through to a different bucket.) The documented recovery for an unreadable outcome — retry, then reconcile — is pinned to be **ledger-safe** rather than merely asserted: against a double that models the engine's own "is this identity already standing?" question, re-sending the same identity comes back as a no-op with the approval ledger and the bucket revision both unmoved, however many times it is repeated, while two concurrent writers each landing a causal point are both honestly reported as having written. Finally the four local gates are pinned to be free: a missing write verb on the injected port, an unwritable direction, an unreadable identity pair, and a principal key that is present but cannot name anyone all refuse **without sending anything** — the last of those because silently degrading a blank target into absence would land a tightening aimed at someone else in the caller's own bucket and return a 200; an empty operator list (501 `capability.operator_principals_required`) is reported as a refusal (`operators_unconfigured`) rather than an unknown outcome |
@@ -424,7 +424,7 @@ guard still cross-checks the table by name).
 | `scripts/run-registry-quota-usage-test.mjs` | The projection of the cloud control plane's quota reading, and the three different things a missing number can mean there. This response says `null` in two places and means something different each time: no token quota is configured for this principal on this instance, and this window has no cap at all. Both are **facts the server is asserting**, not gaps in the reading — while a key that is absent or carries the wrong type is a genuine gap. All three have to survive to the screen separately, because folding them is how a user ends up staring at a confident `0`: an uncapped window rendered as if nothing were left, or a deployment that simply never configured quotas rendered as if the quota were exhausted. The complaint that started this was the opposite direction — a centrally configured quota that the command line could not see at all — so the reading also refuses to let an unreadable response masquerade as "no quota configured". Two fields deliberately do not share one signal: whether a window is exhausted and whether there is a recovery time, since the recovery time is only ever populated in the exhausted case and reading its absence as "not exhausted" would answer a question the response never answered. Counts that the server always provides are narrowed no further than the mint: a used counter has no uncapped state, so a null there is unreadable rather than zero. The wording helper carries the only human-facing phrasing, and the sentences for "no cap" and "unknown" are checked to contain no digits at all |
 | `scripts/run-file-history-capture-capability-test.mjs` | The engine's file-history-capture self-description (`capabilities.fileHistoryCapture`), read the same four-state way as its sibling capability readers: an absent key is reported as not reported (never folded into `off`), words are taken as an open set so a newer mode is not mistaken for a malformed answer, `fileHistoryCaptureMode` recognises `off` and `on-optional` (captured, and a run may opt out) — any other word, including the older `on-always` that pre-`7.104` engines sent, reads as unknown — and the wording for `off` speaks about capture only — whether code can be rewound is left to the rewind readings; a malformed reading makes neither the verdict nor the wording throw. The request side is covered too: the fragment helper yields the per-run opt-out word only when the reading is `on-optional` and the host asked not to capture — never for an older engine that would refuse the unknown key — and the fragment is driven through the real request constructor on every lane. The recognised words are checked both ways against the capture vocabulary of a `7.104` engine package. |
 | `scripts/run-model-identity-resolvability-test.mjs` | The model-identity judgement a client makes before letting anyone in: can the engine it is about to use start with a model name? Each end reports what it read from each place that can feed a model name to a local engine (complete, partial — a gateway address or a credential but no model name —, absent, or unreadable), or, for an engine that runs elsewhere, whether that engine has been seen answering; `modelIdentityResolvability` answers resolvable, unresolvable or unknown. Having part of an upstream configuration is not having enough of one, so partial lanes never add up to resolvable; a lane that was not reported or could not be read makes the answer unknown rather than unresolvable; an engine that runs elsewhere is never judged unresolvable and local lanes are never consulted for it (it does not start without a model name, so seeing it answer is enough to call it resolvable). `modelSetupDecision` combines that answer with whether this end can configure a model at all: setup is offered only for unresolvable on an end that can configure one, an end that cannot says so and points at whoever runs the engine, and unknown never opens setup. The detail and notice sentences are checked to be pairwise distinct, unknown sentences neither claim a model is configured nor that it is not, and the module is checked to import no platform I/O. The catalog lane counts as complete only when the host reports that the local engine accepts a catalog default; a report of `false` reads the catalog as partial, and no report (an explicit `undefined` included) or an unusable one leaves that lane undetermined — the same answer as before the catalog could be complete. Every sentence about a complete catalog names the catalog, including the undetermined sentence when the report is `false` and another lane cannot be read, and a 25-cell matrix pins that the report changes nothing when the catalog is not complete. |
-| `scripts/run-cloud-effective-projection-test.mjs` | The cloud control plane's effective-configuration response beyond its four configuration domains, and what a locally started engine does with the models document derived from it. Three top-level keys are read with the meaning their producer gives them: `warnings` (degradation warnings from the build that produced the served view — an empty list is a clean build, an absent key is an older server that cannot tell), and `budget` / `runtimeCaps`, where `null` means two different things: nothing resolves for this principal when you view yourself, and values withheld when the response previews another principal. A missing key or a wrong type is a third state, unknown, and none of the three is ever folded into a zero, a `false` or "no budget". A malformed warning row, budget field or cap costs only itself, and a known budget field of the wrong type is named as not shown rather than silently read as "no limit on this axis"; warning kinds are an open set, so a kind this client does not recognise still produces a warning line. The budget and cap readers are reconciled against the installed settings schema. The single wording source puts degradation warnings first and keeps every "not set / withheld / unknown" sentence free of digits, while a zero the server really sent is shown as a zero. On the models side, when the host injects an entry check the models document carries the default model, the tier groups and the active tier group (without the check none of the three is written), and every catalog reference the local engine's schema would reject — a default, role, @-mention entry, tier binding or active group that names something outside the catalog served to this principal — is dropped and recorded, because one dangling reference makes the local engine discard the whole models domain and fall back to its environment catalog; this is proven by reading the produced document with the installed file store. An @-mention allowlist that would be pruned to empty is kept as sent, since an empty list means "everything may be mentioned"; that case is recorded, produces its own warning that a locally started engine will reject the cloud model settings and use its environment catalog instead, and the gate reads the document with the installed file store to confirm exactly that outcome, so the sentence turns red the day the local reader becomes lenient. A per-model budget in which no field could be read is never described as having no limits. Registry annotation keys on model entries (`origin`, `overridesTeam`) are removed before the models document is written: the local engine's schema does not accept them, and a configuration refresh would otherwise be rejected as a whole. A model entry the host-injected entry check rejects is left out of the document and references to it are dropped: the local engine drops such an entry at startup, but a refresh rejects the whole configuration over it, so the gate requires a clean read of the produced document; when no entry passes the check, the catalog is kept as sent and gets its own warning, which the gate proves by reading the document back. The check receives a copy, so it cannot alter what is written. Tier words outside the local schema's closed set are dropped and recorded as unsupported, and the package's tier word list is reconciled against the installed schema in both directions; an active tier group is judged against group names, never model names. |
+| `scripts/run-cloud-effective-projection-test.mjs` | The cloud control plane's effective-configuration response beyond its four configuration domains, and what a locally started engine does with the models document derived from it. Three top-level keys are read with the meaning their producer gives them: `warnings` (degradation warnings from the build that produced the served view — an empty list is a clean build, an absent key is an older server that cannot tell), and `budget` / `runtimeCaps`, where `null` means two different things: nothing resolves for this principal when you view yourself, and values withheld when the response previews another principal. A missing key or a wrong type is a third state, unknown, and none of the three is ever folded into a zero, a `false` or "no budget". A malformed warning row, budget field or cap costs only itself, and a known budget field of the wrong type is named as not shown rather than silently read as "no limit on this axis"; warning kinds are an open set, so a kind this client does not recognise still produces a warning line. The budget and cap readers are reconciled against the installed settings schema. The single wording source puts degradation warnings first and keeps every "not set / withheld / unknown" sentence free of digits, while a zero the server really sent is shown as a zero. On the models side, the document is written in the shape of the installed settings schema: the default model rides `roles.default` and the @-mention policy rides `atMention`. When the host injects an entry check the models document carries the default model role, the tier groups and the active tier group (without the check none of the three is written), and every catalog reference the local engine's schema would reject — a role, @-mention entry, tier binding or active group that names something outside the catalog served to this principal — is dropped and recorded, because one dangling reference makes the local engine discard the whole models domain and fall back to its environment catalog; this is proven by reading the produced document with the installed file store. The @-mention policy is narrowed with the settings contract's own narrowing function: an allowlist pruned to empty becomes "nothing may be mentioned" (never "everything"), the "all" mode is left to the schema default, and a policy this client cannot read is kept as sent with its own warning. The keys the settings schema retired (`default`, `atModelAllowlist`) are never written; when an older control plane still sends them, each is recorded and produces its own warning. Warning kinds with their own sentence are reconciled against the kinds the installed schema can produce. A per-model budget in which no field could be read is never described as having no limits. Registry annotation keys on model entries (`origin`, `overridesTeam`) are removed before the models document is written: the local engine's schema does not accept them, and a configuration refresh would otherwise be rejected as a whole. A model entry the host-injected entry check rejects is left out of the document and references to it are dropped: the local engine drops such an entry at startup, but a refresh rejects the whole configuration over it, so the gate requires a clean read of the produced document; when no entry passes the check, the catalog is kept as sent and gets its own warning, which the gate proves by reading the document back. The check receives a copy, so it cannot alter what is written. Tier words outside the local schema's closed set are dropped and recorded as unsupported, and the package's tier word list is reconciled against the installed schema in both directions; an active tier group is judged against group names, never model names. |
 | `scripts/run-websearch-verdict-test.mjs` | The per-request web-search configuration a host puts on the wire. Newer servers refuse the whole request when that section is malformed — and a missing or misspelled search provider now counts as malformed, because the section names where the searches go and an unknown destination is refused rather than silently swapped for the deployment's own backend. The old readers in this package dropped such a section without a word, which let the server swap destinations after all. The guard pins the new three-way verdict (absent, honoured, malformed with the field that is wrong) against the server's own judge, vector by vector, whenever that judge is available next to this package; it pins that a half-configured environment is malformed rather than ignored, that a malformed environment never falls back to the settings file (that would change the destination too), and that neither the sentence shown to the user nor the recorded reason repeats an endpoint, a key, or a search-parameter name or value — an unrecognised provider is never echoed either (the sentence lists the valid words instead), so a URL or key pasted into the wrong field does not come back out, even when it happens to be all letters. A host's key store is plugged in through a callback the package calls only after the provider has been recognised, so the precedence between environment and settings stays inside the package. Since 0.83.0 an endpoint that carries a user name or password is malformed as well (the server judges the same way from 7.101.0), the reason sentences match the server's own word for word, and the three older readers that dropped a misspelled provider are gone. Since 0.93.0 the search key is checked in the same pass: a key that cannot be carried as an HTTP header value — a line break, a NUL or another control character, or a character above U+00FF anywhere but at the very ends — is malformed, with the server's own sentence (the server refuses such a key from 7.118.0 on, and a deployment's own start-up configuration is refused too), while a trailing or leading newline, as secret files often carry, is accepted just as the runtime accepts it. The rule is a pure re-implementation of the runtime's two-step header check, so the package stays free of Node built-ins; the guard pins it character by character against the real Node header pipeline, sends the whole Latin-1 range through a real loopback request, compares a large corpus with the previous release byte for byte (legal keys read exactly the same; only keys that could never be sent differ), and pins that the server's refusal is read by its code alone — nothing in the package matches its sentence. |
 | `scripts/run-hooks-merged-disable-projection-test.mjs` | The fourth governance leg of the hooks projection: `disableAllHooks` set in a non-managed settings source. The value that counts is the **merged** one, read from the host through the optional `SettingsPort.mergedDisableAllHooks()`, because a per-source approximation ("any source says true") reads user `true` with local `false` backwards — the merged value there is `false` and every source's hooks ship. When the merged value is `true`, only managed-settings hooks are sent to the engine: non-managed settings can switch off their own hooks, never the managed ones, and a managed `disableAllHooks` is still judged first and sends nothing at all. A full matrix over the four sources, each true, false or absent, is merged with the reference rule (later sources override earlier ones, managed settings last) and every cell's projection is asserted. The reader is the only authority: per-source values never second-guess it, and only a strict `true` counts. A host that does not implement it keeps the previous behaviour and gets exactly one warning per installed settings port, never one per request, and none on paths where the reader would not have been consulted; a reader that throws is treated as `true`, so managed hooks still ship. The session goal's Stop hook and the final-verification yield rule, which reads the projected hooks, follow the same verdict, and the trust gate and the three managed gates are evaluated before the reader is ever called. The last leg pins the member's declared shape in the built declarations: optional, no parameters, returning a boolean or `undefined`. The flag settings source (a settings file or inline settings given at startup) is projected after the local source, and settings hooks stay concatenated managed first (managed → user → project → local → flag): the engine runs hooks in request order, and three of its budgets go to whoever comes first — the per-event wall-clock budget, the per-event cap on model-backed entries, and the total cap on added context — so a managed hook placed after the others could be crowded out and silently not run. The flag source moves with user, project and local under every managed or merged gate; its exec-form entries are dropped and warned about like any other settings source, and the not-run notice names it `Flag settings`. An optional leg feeds the request body to an installed server's hook runner and requires the managed deny, block and context to take effect while non-managed hooks exhaust each of those budgets, and pins the known cost of that order: a non-managed `PreToolUse` hook placed after the managed one can still rewrite the input after the managed check allowed it. Without an installed server the leg reports that it did not run. Safe and bare mode: the host reports its startup mode through the optional `SettingsPort.hooksStartupMode()`, which both the plain and the plan path read and which answers `'safe'` or `'bare'`; the flag already carried by the plugin hooks reading, which only the plan path reads and which cannot tell the two modes apart, counts as safe mode unless the reader named a mode. In either mode only managed-settings hooks are sent from the settings sources: bare mode is handled like safe mode, so managed hooks are still sent, also under the managed-only gates. In both modes plugin hooks are excluded as before and the session goal's Stop hook is still sent. Only the two exact words count; a host that does not implement the reader, returns `undefined`, returns any other value or throws keeps the previous request byte for byte, checked for every such reply, both flag states, both paths and with or without a session goal; any other value and a throw each leave one debug line, and so does a reader written as a property instead of a method, which narrows nothing. The reader is called at most once per request and not at all when hooks are already switched off entirely. The optional server leg requires the left-out sources' hooks not to run while the managed ones still run and still deny, in either mode. A hook written in more than one settings place is sent once: entries on the same event, in groups that are identical apart from their hooks (same matcher), with the same identity — command, shell, arguments and condition for command hooks; type, prompt and condition for prompt and agent hooks; URL for HTTP hooks; server, tool and input for MCP tool hooks — are sent once, at the position of their first occurrence in the concatenation order and with the fields of the copy from the highest-precedence source (managed, then flag, local, project, user; within one source the later copy), so a managed copy keeps both its place and its values and a user copy duplicated in flag settings keeps its place but uses the flag copy's timeout; the other entries keep their order, groups without duplicates are sent as they were, and a group left empty is not sent. Different matchers, events, a one-byte command difference, a different condition, an absent versus explicit shell, different types or extra group keys are not merged; malformed entries and entries whose identity cannot be computed are left alone; plugin groups and the session goal hook do not take part, and hooks removed by safe or bare mode never come back. The optional server leg requires such a duplicate to run once with its added context appearing once, and a user copy with a one-second timeout duplicated by a flag copy with a ten-second timeout to run to completion on a two-second command. |
 | `scripts/run-memory-saved-projection-test.mjs` | Engine memory writes (a successful `Remember` tool call) moved off the transcript onto the additive `memory_saved` chrome event, driven through the real pipeline: zero transcript rows for the write (the transcript is byte-identical to the same frames with the write reported as not successful), exactly one event whose `notes` carry the note text verbatim (notes, not file paths) and whose key set is exactly kind / laneProof / id / notes; no event for a missing, empty or non-string note, a non-`true` `ok`, a tool error, a missing result or another tool name; two writes give two events in order with distinct ids; a sub-agent write rides the sub-agent lane and an empty parent id emits nothing rather than falling back to the main lane; the `id` is derived from the write's wire key (the tool-end event id, else the tool-start event id, else the call id; empty ids count as absent), so projecting the same wire events twice gives the same id, and it never collides with the tool result row of the same or another call; the event sits right after the tool result row; the arm is registered as required. |
@@ -432,9 +432,9 @@ guard still cross-checks the table by name).
 | `scripts/run-layering-shadow-export-test.mjs` | Same-name shadows across the first-party clients that consume this package (terminal, desktop, web and the admin console). Each client's product sources are read at the local clone's `origin/main` (or its HEAD when there is no such ref), without fetching, and parsed with the TypeScript parser; every top-level runtime export the client declares itself is compared with this package's public runtime exports. The guard prints which ref, commit and commit date it read for each client, and warns (without failing) when that commit is more than seven days old, because the result then only describes that older snapshot. A client-side declaration carrying the name of a package export means a piece of shared logic now lives in two places and can drift apart. It fails the guard unless it is listed in `scripts/layering-shadow-exemptions.json`, and a listed row must carry one due-by version no more than three minor lines ahead (it fails once the package reaches it; that single field is the only one the guard reads for expiry, and the older retire-by and review-by fields that still sit on the rows for one more version are only checked to agree with it). It also fails once the client has removed the shadow and the row still stands. Every row also names what kind of duplicate it is, and carries the evidence that kind requires. A client copy that is this package's own function object behind a type assertion, or a thin wrapper that only passes the client's own dependencies into this package's function, must be proven so on every run by reading the client's syntax tree; such rows are due for review within two minor versions, and the guard fails as soon as the proof no longer holds. A wrapper whose extra logic the client has confirmed to be host-specific carries that confirmation (the client's own wording and where it was stated) together with a pinned review record that is due within two minor versions. A wrapper that adds its own decisions, or a same-name function that does something else, carries a review record pinned to a hash of the normalized client declaration (comments and formatting do not count); once the client's copy changes, the guard fails until it is reviewed again. A second implementation of the same logic carries a semantic diff: the client's copy is taken from the commit being read, the named declarations and the client modules they import are extracted with the TypeScript parser, transpiled and run in memory against this package's build on the same inputs, and any disagreement outside the classes registered on that row fails the guard (each class is a fixed predicate over both answers). A separate register lists same-meaning copies that carry a different name (a name-based comparison cannot see those): each row names the client declaration and the package export it duplicates, and fails when it is malformed, past its due-by version, or when the client no longer declares that name; the register only records and checks liveness and expiry, it does not scan for such copies. A row can follow a client-side rename to a new name, and can be registered ahead of a client change for a limited number of versions before the client code exists. Re-exports of this package's own exports are the intended form and never count. A client tree that is not present is reported as a skipped section, not as a pass. The ruler proves itself on an in-memory fake client (planted shadows must be caught, legal forms must not), on a throwaway repository (a missing `origin/main` falls back to HEAD, a broken one is a fault rather than a silent fallback), and refuses to report zero on a client whose scan surface is empty. |
 | `scripts/run-session-policy-deliverable-test.mjs` | Which of a batch of user-written permission rules can be written into a session’s own rule record without changing their meaning, and why each of the others cannot. The record holds whole tool names and command names only, so exactly one class maps across losslessly: a deny rule that names one tool with no qualifier. Everything else is withheld with one word from a closed seven-word list — an ask rule (the record has no ask tier), a deny rule with a parenthesised qualifier (recording just the name could block more), a rule that names a server or agent peer without naming one of its tools (for every protocol namespace the engine knows, checked against the engine package's own table) or contains a wildcard (*) anywhere (an engine that compares exact names would block nothing), an entry that is not a tool name, and a name the engine refuses at the start of every run — a retired tool name, or one containing "__" without a protocol prefix, where the prefix check is case-sensitive (once such a name is in the record, every later run of the session fails at startup until that entry is removed; the retired-name list is checked entry by entry against the engine package's own list, and a withheld retired name carries its current name when the engine says it was renamed), and — only when the caller passes the tool roster of a run — a name that roster does not list as a tool name or alias, compared exactly with letter case counted (recorded as it is it would block nothing on that deployment) — and each word has one sentence, which never echoes the rule itself; asking for the sentence never throws, even with a value that throws when turned into a string. A name with leading or trailing whitespace counts as not a tool name: the record compares exact bytes, so it would block nothing. The guard pins the batch semantics: the deliverable part is either the whole batch or empty, never a subset, so a caller cannot send half a change and report it as saved. An end-to-end check runs the engine package itself: every batch this function would deliver — the recorded vectors and a fixed-seed sample of generated names — is written into an in-memory session rule store and the next run must get past its start-up checks and reach the model, while every name withheld as refused — every retired name included — must indeed make that run fail at start-up, and every string literal in the judgement source that it withholds as refused must be one the engine package's own tables refuse. It also checks that malformed input never throws and never delivers anything (non-arrays, non-string entries, holes, a polluted array prototype, a length or index that throws, a changing index read once), that a batch which cannot be read at all is marked `unreadable: true` while an empty batch is not, so the two stay tellable apart, that each word is produced by some vector and nothing outside the list is produced, and — when a checkout of the previous in-client implementation is present — that this function gives the same answer on every recorded vector and on tens of thousands of generated rules and pairs, except for five deliberately stricter classes (whitespace-padded names; rules with a wildcard anywhere, which the previous implementation sent as exact names unless the wildcard was the whole tool part of a server rule; peer-wide rules outside the MCP namespace, which it did not recognise; names the engine refuses at start-up, which it sent as ordinary names; and names the engine's rule compiler does not accept as one whole tool, such as an escaped parenthesis or a backslash in the name), whose disagreements are counted per class and must match an independent count exactly. The optional roster only ever withholds more: with no roster, or an absent or null one, every answer is byte-identical to the last release before the roster existed, checked against that release's published file over tens of thousands of inputs, apart from the rule-syntax class described below; a name withheld without a roster stays withheld with the same word, and the same current name, when the roster lists it as a name or an alias; a name delivered without a roster is withheld as not in the roster exactly when the roster lacks it; the engine's own namespace-covering forms keep their word; and a roster or options value that cannot be read marks the whole batch unreadable instead of falling back to no roster, while a readable empty roster is a roster. An end-to-end check takes the roster from a real run of the engine package that mounts a tool with an alias: every name delivered with that roster passes the engine's start-up name audit without being reported, and every name withheld as not in the roster is one that audit reports as matching no mounted tool. When the previous in-client implementation carries its own roster check, the two are compared under a given roster and may differ only for an alias (delivered here, withheld there) and for a refused name that the roster lists (withheld here, delivered there). A deny rule is delivered only when the installed engine's rule compiler, asked about that one string the same way the request-body permission assembly asks about whole tool names, accepts it as a rule for one whole tool: a name with an escaped parenthesis, a backslash, a lone closing parenthesis or more characters than a rule may have is withheld as not a tool name, because the record would compare those bytes and match no tool. When the compiler is not installed or cannot answer, a batch that needs it is marked unreadable rather than delivered, while batches that never reach that step answer as before. The guard checks that every delivered name is one the compiler accepts, that nothing delivered now was withheld by the earlier release and the extra withholding is exactly that class, that a roster never makes such a name deliverable, that the compiler of every engine generation in the server fixtures gives the same answers, and that the request-body assembly reaches the same verdict on the same names. |
 | `scripts/run-plugin-hooks-projection-test.mjs` | Plugin hooks: each command hook an enabled plugin declares is decided one by one as running in the engine, running in this client, or not running at all, and the page of hooks sent with a request is built from the same per-turn plan the client uses to skip its own copies, so one hook never runs in two places. Governance is judged first and always wins — a managed hooks switch-off, an untrusted workspace, safe or bare mode, or a governance read that fails sends no plugin hook and does not list it as a gap; managed-hooks-only (set directly, or through a merged non-managed hooks switch-off) keeps only managed plugins; the plugin-only customization lock does not touch plugin hooks. A hook reaches the engine only when this client started the engine on this machine, the engine reports plugin-hook support, the entry is a command, the plugin declares no sensitive option, and the event still fits the engine's per-event limits; the gate walks that matrix cell by cell, including the limit boundaries and a session goal hook counting toward them. A fact that was never read is reported as not known rather than as a fact: a host that does not say where the engine runs gets a "not known whether this client started the engine" reason, an engine whose capabilities have not been read yet gets a "not known yet whether it supports plugin hooks" reason, and the plan's two engine facts are null in those cases, not false. Events the engine never fires run only if the client says it fires them itself, and hooks the upstream behaviour itself refuses (option references in a shell-form command, an unset option in exec form, malformed entries) run nowhere. Exec-form arguments are passed element by element with only saved non-sensitive option references filled in; path placeholders are left for the executor. Sensitive option values never reach the request: with a host that wrongly supplies one, every string in the plan, the request body, the notice, the labels and the log is searched for it across eight cells. A host without the plugin reader keeps the previous request body and gets exactly one warning per settings port; plugin data that throws while it is being read (a throwing getter, a revoked proxy) is treated like a failing reader — no plugin hooks this turn, settings hooks still sent, nothing thrown; the not-running notice names the plugin and events, never a command or an option value, and escapes control characters in names. Command hooks from settings that carry arguments (a non-empty `args` array, which is the exec form, or any other non-null value) are removed from the request until the engine reports support for arguments, because the engine would otherwise drop the arguments and run the bare command through a shell; an empty `args` array is not treated as carrying arguments when the command is made only of letters, digits and `_ . / : + -` (the shell runs the same executable), so such a guard still reaches the engine, while an empty array on a command with spaces or shell characters is removed; `args` on a prompt or http entry, a null `args`, or an entry with no type is left alone, and those go out unchanged. MCP tool hooks, which the engine cannot parse, are removed only from a request built from a plan, whose not-running notice the host shows; a request built without a plan still carries them on engine-fired events, so the engine rejects the whole request loudly instead of a guard hook silently not running — the gate checks both request bodies against the engine's own hooks schema. Without a plan, every removed hook of that kind on an engine-fired event produces one warning per settings port, event and reason. Malformed entries still pass through for the engine to reject loudly, and passing null where the options object goes behaves like passing nothing; a `plan` option that is not a plan is ignored rather than turning the whole page into nothing, and a plan passed directly in place of the options object is recognised and used. A `plugin` key written by hand on a settings hook is stripped before sending (even when its value is undefined), because only hooks that come from the plugin reader may carry plugin context; the settings document itself is left untouched and a debug line records the count. The two hand-copied tables, the engine-fired event list and the engine limits, are checked against their owners. Since 0.89.0 the same per-turn decision also gives every settings hook on an engine-fired event exactly one executor: a command, HTTP, prompt or agent hook from settings goes to the engine when the engine reports that it runs hooks sent with requests and either this client started the engine on this machine or this client cannot run that hook itself; otherwise this client runs it and the hook is left out of the request. A hook the engine is known to skip (a PowerShell command, or hooks the engine ignores entirely) that this client cannot run either is listed as not running; one the engine may or may not reach (capabilities not read yet, or a prompt or agent hook after the fourth one on an event) is still sent and not listed. The client learns which of its own settings hooks to skip from `settingsHooksKeptLocally`, which matches hooks by content against the request body only, returns the same array when nothing is skipped, and for the two compaction events follows whether the engine actually compacted. When the prompt was not written by the user, prompt-submit hooks from settings and plugins are left out on both sides. A host that reports events must also report which hook types its own executor runs on each event: with that table, a plugin hook it cannot run is reported as not running rather than as running in this client; without it, prompt and agent hooks from settings on engine events are treated as hooks the client cannot run and stay in the request for the engine. A settings hook whose required field is unusable (a command, URL or prompt that is not a non-empty string) is not assigned to either side and is sent unchanged, so the service still rejects it; malformed rows in a plan handed back by the host never make the not-running notice throw and never authorise skipping a hook; and a missing, empty or `*` matcher counts as the same match-all matcher when the client decides what to skip. A matrix over source, event, hook type, capability, engine ownership, host shape, prompt origin and compaction outcome checks that every hook runs at most once and exactly once whenever either side can run it, while the print lane and hosts without a local executor keep the previous request byte for byte. An optional leg runs the request body through an installed server's hook runner, current and previous release, together with the client-side skip, and counts one run per prompt. A plan in the previous release's six-key shape is read the same way by the request side and the skip side, so every hook it sends runs once; the client side of the skip reads groups and entries the way the client's executor does (class instances and inherited getters included); and on the two compaction events a prompt or agent hook past the engine's per-event limit runs in the client when the client can run it. |
-| `scripts/run-display-untrusted-projection-test.mjs` | The single display-safety outlet (`displayUntrusted`) and the credential wash on the end-of-run rows this package mints. The outlet composes two credential nets (URL structure: userinfo, every query value, the fragment, path parameters and path segments that start with a known secret prefix; key/value words such as `Authorization: Bearer ...`, `Authorization: token ...` or `api_key=...`, plus well-known secret literals that appear without a label, such as `sk-...`, `ghp_...`, `AKIA...`, JWTs and the body of a PEM private key) with three character nets (control characters, bidirectional and format characters, whitespace folding). The credential nets match on a view of the text with ANSI sequences, format characters, control characters and the outlet's own escape tokens stripped, and map the result back onto the original, so colouring or an invisible character wedged between a label, its separator and its value cannot hide the value, and no stray marker is left behind. Whitespace of any length around the separator is accepted. Hosts, ports, paths, query key names and surrounding prose stay byte-for-byte, clean text comes back unchanged, the result is idempotent (also with a length cap), a length cap never splits an escape token or a surrogate pair, an invalid cap means no cap, and every net can be switched off on its own. A few narrow shapes are left alone because they name something rather than carry a value (a plain English word after `bearer` or `basic`, a back-quoted credential variable name, a plain integer after `tokens:`, a list of key names after `keys:`), each with a counter-example that is still washed. Regional flag emoji built from tag characters are kept whole. The existing single-line helpers (`escapeDisplayControlChars`, `collapseLabel`, `capForDisplay`, peer sender names and the hook failure banner) now run on the same engine and are held byte-identical to their previous output over every BMP code unit plus random strings. The approval decision-note echo, the subagent resume receipt (and its failure debug line) and the startup list of plugin hooks that will not run now also drop bidirectional and format characters (and, for the receipt, C1 controls); a note that is empty after cleaning is treated as absent. The synthetic end-of-run rows (`API Error:`, `Run stopped:`, `Model output error:`, `Outcome unknown:`) and the result frame's `errors[]` pass both credential nets before they leave the package, on the print lane and on the interactive lane (which also keeps the row-class flag); this covers a blocked reason whoever wrote it, while assistant text rows, a successful `result` and salvaged output are never touched, and a non-string `errors[]` entry is passed through unchanged. The known-secret-prefix check is a local copy of the configuration package's detector and is compared with the installed one entry by entry. Since 0.83.5 the outlet has two opt-in switches and a position read-out. `escapeBackslashes` (escape form only) writes every literal backslash as a pair, so each output decodes back to exactly one input (a real invisible character and its literal six-character spelling no longer look alike); a reference decoder round-trips thousands of random strings, the output is byte-identical to the default when the input has no backslash, a length cap is measured on the paired output and keeps the longest fitting prefix, credentials are washed exactly as in the default, and the switch is not idempotent by design (use it only at the final render). Zero-width joiners and non-joiners are kept only inside emoji sequences drawn as emoji (so not between symbols such as © or ™ that display as text) and between letters of scripts where they change the shaping (joining scripts such as Arabic, and the Brahmic family), each listed script checked both ways; the one other place a joiner is kept is right after a virama at the end of a word, the older spelling still found in Malayalam and Bengali text. Next to Latin, Cyrillic, CJK and other letters, next to modifier letters shared across scripts, at the start of a word, at the end of a word without a virama before it, or on their own they are now marked. `blanks` marks characters that look like a space but are not an ASCII space (no-break and other width spaces, the ideographic space, Hangul fillers, the blank Braille pattern) before whitespace folding, for names that must never look alike. `displayUntrustedMarks` returns the same text plus the position of every character mark; its text is compared with the outlet over thousands of inputs. With both credential nets off, the character face stays byte-identical to the previous release for input without joiners. The credential nets read escape sequences the way a terminal would when one is cut short: an unfinished colouring or character-set sequence interrupted by another one is dropped as a whole, a sequence never takes the `@` of an address as its final character, and a final character that starts a well-known secret literal (`sk-`, `ghp_`, `AKIA`, a JWT) is also read as the start of that literal; escape tokens this outlet writes are read as one unit, while look-alike text it never writes (an upper-case `\U`, or a code point it never marks) is read as plain text. A URL is cut before a run of non-ASCII blanks followed by a credential label or scheme word, Hangul fillers and the blank Braille pattern count as spaces around a label's separator, and a value that itself starts with a quoted label (`token= "password":"..."`) is left to that inner label. With `blanks` on, a blank written as an escape token right after a label is read as a blank when the value is judged, so `password:` followed by a no-break space and `missing` stays unmasked and an empty value gets no marker; the credential nets also read the text the way it looks after default whitespace folding and combine what each reading masks, so whatever the default form masks stays masked with `blanks` on (checked over a seeded corpus for the escape form, with paired backslashes and without folding; the exceptions are text that itself contains a literal six-character blank escape, which cannot be told apart from one the outlet wrote, and the dot and space marks, which cannot tell a marked blank from a real dot or space). A lone surrogate wedged between a label, its separator and its value no longer hides the value: the credential nets treat it exactly like a format character, in the machine-readable wash, in a single pass of the display outlet, and on the end-of-run rows and the result frame's `errors[]` on both lanes, while lone surrogates anywhere else are left byte-for-byte. Since 0.84.1 an address-shaped value (`scheme://…`) that sits right after an invisible single character — a format character, a lone surrogate, a non-whitespace control character, or the outlet's own escape token for one — is no longer let through as an address: the scheme stays and the host and path are replaced, with the query and fragment masked as before; `user:<password>@` followed only by stripped units and then a boundary, a port or another `@` is masked as userinfo. A value after real whitespace or after a colour sequence is still treated as an address, and clean text stays unchanged; forms that the previous release masked are checked not to leak on the same outlets. A `user:<password>@` candidate that is the value of a credential label or scheme word — including a label or scheme word split by stripped characters, and a quoted value that goes on past whitespace — is left to the label pass, so the whole value is masked as in the previous release; both reported shapes and frozen samples from the targeted pools are checked on every outlet. A scheme-less `user:password@host` whose password holds a double quote, backtick or angle bracket is still masked when a credential label or scheme word sits right before the user name, or when the user name is a plain word that is not itself a label, the host has a real shape and the address is not wrapped by that quote; quoted JSON values, quoted or bracketed e-mail addresses and quoted arrays stay byte-identical. When a label inside a quoted value ends right at that value's closing quote, the quote and separator stay outside the mask, and an address that begins the rest of a quoted value counts as an address only up to the next blank, so washing the output again never changes it. The credential stage now repeats its single scan on its own output until nothing changes (the cap counts rewrites, at most eight; clean text takes exactly one pass), so masking the output again never changes it for every input that settles within the cap — the random, exhaustive and grammar pools and the two known step-by-step constructions settle in at most three passes counting the confirming one, and the searched constructions kept in the pool in at most six; a deliberately constructed input can need more rewrites as it grows, and at the cap the last rewrite is returned, which a further pass can only mask more of. The checks cover the pass cap and how reaching it is reported, single-pass clean text, a grammar sample of the busiest shapes, constructions that once advanced one step per pass, the mark positions on the final output and how positions compose across rewrites. A username that is itself a credential label with a quote or angle bracket in the password is masked as userinfo (also with only the URL net on) unless the label's own value really reaches the `@`; a URL parameter such as `;token=...` or `,token=...` stays inside the URL piece; a private-key header the outlet cannot recognise does not hide the labels inside its block; the tail glued after the host of a label's address value is masked to the value's end; and a value this package's own transcript wash split at a carriage return is masked when the display joins it back. A bare `Token <value>` / `token <value>` (no label, no separator) is now washed like `Bearer`, with a wider allow-list of ordinary English words after `token` so prose such as `token count: 1234` or `token expired` stays readable; a `/token` URL path tail followed by a status code is not a scheme word and stays as is, and neither is `token` right after `unexpected` in a parser error such as `SyntaxError: Unexpected token '<'`, which stays readable. The `token` scheme word only adds masking of its own value: label values and `bearer` / `basic` values are read exactly as in the previous release, so a value that starts with another character, or a `bearer` value, followed by `token`, a non-ASCII space and an address is still masked whole. When a credential label or scheme word is followed by a colon and a YAML block scalar header (`\|` or `>`, optionally with `-`, `+` or digits, invisible characters allowed between them, and nothing else on that line but blanks, invisible characters or a comment, ending in a line break), the whole indented block becomes a single marker: the block runs until the first line with visible content that is not indented deeper than the label's line, blank lines included, and the line break before that line and the line itself stay as they were on outlets that keep line breaks (outlets that draw line breaks as escapes or dots merge the next word into the marker, as they already do after any value). Raw and JSON-escaped line breaks are both read; blanks inside escape sequences or control strings never count as indentation, and a block line holding only a control string counts as content and is masked whole. Forms where the marker is followed by more text on the same line, `=`-separated forms such as `key = \|` or the `=>` arrow, and a header at the very end of the text read exactly as in the previous release. From 0.94.0 the private-key header has one definition, generated from the settings contract: the whole-string check finds it anywhere in the text (as the contract's own write-side check does from its 10.0.0 release), and the key-block cutter builds its header line from the same definition, so a lowercase header or one without a space after BEGIN has its key body masked like the canonical form, while public keys and certificates are left as they were. Since 0.94.0 the credential words also read through YAML node properties (type tags such as `!!str` and anchors such as `&name`) between the colon and the value: the properties and the value after them get one marker each, block scalars and block-sequence items included, and a property with no value after it on the same line reads exactly as before. A list after a credential label (a block sequence, a flow sequence including JSON arrays, or a same-line comma list after a plural label) is judged item by item with the same exemptions as a single value: a list of plain key names after a bare `keys`, or of integers after a bare `tokens`, is left alone, and any value-like item masks every non-neutral item. A dash right after the colon on the same line is a list bullet, not a value. The retry overlay's error text (`error.formatted` on the four error arms) passes through the same outlet, and the label vocabulary is pinned as it stands (members masked, known out-of-family keys such as `passphrase` left as written). |
+| `scripts/run-display-untrusted-projection-test.mjs` | The single display-safety outlet (`displayUntrusted`) and the credential wash on the end-of-run rows this package mints. The outlet composes two credential nets (URL structure: userinfo, every query value, the fragment, path parameters and path segments that start with a known secret prefix; key/value words such as `Authorization: Bearer ...`, `Authorization: token ...` or `api_key=...`, plus well-known secret literals that appear without a label, such as `sk-...`, `ghp_...`, `AKIA...`, JWTs and the body of a PEM private key) with three character nets (control characters, bidirectional and format characters, whitespace folding). The credential nets match on a view of the text with ANSI sequences, format characters, control characters and the outlet's own escape tokens stripped, and map the result back onto the original, so colouring or an invisible character wedged between a label, its separator and its value cannot hide the value, and no stray marker is left behind. Whitespace of any length around the separator is accepted. Hosts, ports, paths, query key names and surrounding prose stay byte-for-byte, clean text comes back unchanged, the result is idempotent (also with a length cap), a length cap never splits an escape token or a surrogate pair, an invalid cap means no cap, and every net can be switched off on its own. A few narrow shapes are left alone because they name something rather than carry a value (a plain English word after `bearer` or `basic`, a back-quoted credential variable name, a plain integer after `tokens:`, a list of key names after `keys:`), each with a counter-example that is still washed. Regional flag emoji built from tag characters are kept whole. The existing single-line helpers (`escapeDisplayControlChars`, `collapseLabel`, `capForDisplay`, peer sender names and the hook failure banner) now run on the same engine and are held byte-identical to their previous output over every BMP code unit plus random strings. The approval decision-note echo, the subagent resume receipt (and its failure debug line) and the startup list of plugin hooks that will not run now also drop bidirectional and format characters (and, for the receipt, C1 controls); a note that is empty after cleaning is treated as absent. The synthetic end-of-run rows (`API Error:`, `Run stopped:`, `Model output error:`, `Outcome unknown:`) and the result frame's `errors[]` pass both credential nets before they leave the package, on the print lane and on the interactive lane (which also keeps the row-class flag); this covers a blocked reason whoever wrote it, while assistant text rows, a successful `result` and salvaged output are never touched, and a non-string `errors[]` entry is passed through unchanged. The known-secret-prefix check is a local copy of the configuration package's detector and is compared with the installed one entry by entry. Since 0.83.5 the outlet has two opt-in switches and a position read-out. `escapeBackslashes` (escape form only) writes every literal backslash as a pair, so each output decodes back to exactly one input (a real invisible character and its literal six-character spelling no longer look alike); a reference decoder round-trips thousands of random strings, the output is byte-identical to the default when the input has no backslash, a length cap is measured on the paired output and keeps the longest fitting prefix, credentials are washed exactly as in the default, and the switch is not idempotent by design (use it only at the final render). Zero-width joiners and non-joiners are kept only inside emoji sequences drawn as emoji (so not between symbols such as © or ™ that display as text) and between letters of scripts where they change the shaping (joining scripts such as Arabic, and the Brahmic family), each listed script checked both ways; the one other place a joiner is kept is right after a virama at the end of a word, the older spelling still found in Malayalam and Bengali text. Next to Latin, Cyrillic, CJK and other letters, next to modifier letters shared across scripts, at the start of a word, at the end of a word without a virama before it, or on their own they are now marked. `blanks` marks characters that look like a space but are not an ASCII space (no-break and other width spaces, the ideographic space, Hangul fillers, the blank Braille pattern) before whitespace folding, for names that must never look alike. `displayUntrustedMarks` returns the same text plus the position of every character mark; its text is compared with the outlet over thousands of inputs. With both credential nets off, the character face stays byte-identical to the previous release for input without joiners. The credential nets read escape sequences the way a terminal would when one is cut short: an unfinished colouring or character-set sequence interrupted by another one is dropped as a whole, a sequence never takes the `@` of an address as its final character, and a final character that starts a well-known secret literal (`sk-`, `ghp_`, `AKIA`, a JWT) is also read as the start of that literal; escape tokens this outlet writes are read as one unit, while look-alike text it never writes (an upper-case `\U`, or a code point it never marks) is read as plain text. A URL is cut before a run of non-ASCII blanks followed by a credential label or scheme word, Hangul fillers and the blank Braille pattern count as spaces around a label's separator, and a value that itself starts with a quoted label (`token= "password":"..."`) is left to that inner label. With `blanks` on, a blank written as an escape token right after a label is read as a blank when the value is judged, so `password:` followed by a no-break space and `missing` stays unmasked and an empty value gets no marker; the credential nets also read the text the way it looks after default whitespace folding and combine what each reading masks, so whatever the default form masks stays masked with `blanks` on (checked over a seeded corpus for the escape form, with paired backslashes and without folding; the exceptions are text that itself contains a literal six-character blank escape, which cannot be told apart from one the outlet wrote, and the dot and space marks, which cannot tell a marked blank from a real dot or space). A lone surrogate wedged between a label, its separator and its value no longer hides the value: the credential nets treat it exactly like a format character, in the machine-readable wash, in a single pass of the display outlet, and on the end-of-run rows and the result frame's `errors[]` on both lanes, while lone surrogates anywhere else are left byte-for-byte. Since 0.84.1 an address-shaped value (`scheme://…`) that sits right after an invisible single character — a format character, a lone surrogate, a non-whitespace control character, or the outlet's own escape token for one — is no longer let through as an address: the scheme stays and the host and path are replaced, with the query and fragment masked as before; `user:<password>@` followed only by stripped units and then a boundary, a port or another `@` is masked as userinfo. A value after real whitespace or after a colour sequence is still treated as an address, and clean text stays unchanged; forms that the previous release masked are checked not to leak on the same outlets. A `user:<password>@` candidate that is the value of a credential label or scheme word — including a label or scheme word split by stripped characters, and a quoted value that goes on past whitespace — is left to the label pass, so the whole value is masked as in the previous release; both reported shapes and frozen samples from the targeted pools are checked on every outlet. A scheme-less `user:password@host` whose password holds a double quote, backtick or angle bracket is still masked when a credential label or scheme word sits right before the user name, or when the user name is a plain word that is not itself a label, the host has a real shape and the address is not wrapped by that quote; quoted JSON values, quoted or bracketed e-mail addresses and quoted arrays stay byte-identical. When a label inside a quoted value ends right at that value's closing quote, the quote and separator stay outside the mask, and an address that begins the rest of a quoted value counts as an address only up to the next blank, so washing the output again never changes it. The credential stage now repeats its single scan on its own output until nothing changes (the cap counts rewrites, at most eight; clean text takes exactly one pass), so masking the output again never changes it for every input that settles within the cap — the random, exhaustive and grammar pools and the two known step-by-step constructions settle in at most three passes counting the confirming one, and the searched constructions kept in the pool in at most six; a deliberately constructed input can need more rewrites as it grows, and at the cap the last rewrite is returned, which a further pass can only mask more of. The checks cover the pass cap and how reaching it is reported, single-pass clean text, a grammar sample of the busiest shapes, constructions that once advanced one step per pass, the mark positions on the final output and how positions compose across rewrites. A username that is itself a credential label with a quote or angle bracket in the password is masked as userinfo (also with only the URL net on) unless the label's own value really reaches the `@`; a URL parameter such as `;token=...` or `,token=...` stays inside the URL piece; a private-key header the outlet cannot recognise does not hide the labels inside its block; the tail glued after the host of a label's address value is masked to the value's end; and a value this package's own transcript wash split at a carriage return is masked when the display joins it back. A bare `Token <value>` / `token <value>` (no label, no separator) is now washed like `Bearer`, with a wider allow-list of ordinary English words after `token` so prose such as `token count: 1234` or `token expired` stays readable; a `/token` URL path tail followed by a status code is not a scheme word and stays as is, and neither is `token` right after `unexpected` in a parser error such as `SyntaxError: Unexpected token '<'`, which stays readable. The `token` scheme word only adds masking of its own value: label values and `bearer` / `basic` values are read exactly as in the previous release, so a value that starts with another character, or a `bearer` value, followed by `token`, a non-ASCII space and an address is still masked whole. When a credential label or scheme word is followed by a colon and a YAML block scalar header (`\|` or `>`, optionally with `-`, `+` or digits, invisible characters allowed between them, and nothing else on that line but blanks, invisible characters or a comment, ending in a line break), the whole indented block becomes a single marker: the block runs until the first line with visible content that is not indented deeper than the label's line, blank lines included, and the line break before that line and the line itself stay as they were on outlets that keep line breaks (outlets that draw line breaks as escapes or dots merge the next word into the marker, as they already do after any value). Raw and JSON-escaped line breaks are both read; blanks inside escape sequences or control strings never count as indentation, and a block line holding only a control string counts as content and is masked whole. Forms where the marker is followed by more text on the same line, `=`-separated forms such as `key = \|` or the `=>` arrow, and a header at the very end of the text read exactly as in the previous release. From 0.94.0 the private-key header has one definition, generated from the settings contract: the whole-string check finds it anywhere in the text (as the contract's own write-side check does from its 10.0.0 release), and the key-block cutter builds its header line from the same definition, so a lowercase header or one without a space after BEGIN has its key body masked like the canonical form, while public keys and certificates are left as they were. Since 0.94.0 the credential words also read through YAML node properties (type tags such as `!!str` and anchors such as `&name`) between the colon and the value: the properties and the value after them get one marker each, block scalars and block-sequence items included, and a property with no value after it on the same line reads exactly as before. A list after a credential label (a block sequence, a flow sequence including JSON arrays, or a same-line comma list after a plural label) is judged item by item with the same exemptions as a single value: a list of plain key names after a bare `keys`, or of integers after a bare `tokens`, is left alone, and any value-like item masks every non-neutral item. A dash right after the colon on the same line is a list bullet, not a value. The retry overlay's error text (`error.formatted` on the four error arms) passes through the same outlet, and the label vocabulary is pinned as it stands (members masked, known out-of-family keys such as `passphrase` left as written). From 0.95.0 the credential words also read a value wrapped in round or square brackets after a scheme word or label (placeholder-shaped contents such as `(API_KEY)` or `(your-token)` and bracketed prose stay as written), a JSON-escaped tab between a scheme word or label and its value, a same-line associative-array arrow with a quoted label and a quoted value, and a `token` scheme word with its value after a credential label as one value; a word with no label at all is masked when it is shaped like a random token (length, three character classes, entropy and run structure), while hex hashes, UUID lists, base64 image headers and data URIs, integrity strings, provider object ids and ordinary identifiers stay as written; a known secret prefix glued after an underscore or hyphen inside an identifier or a URL path segment is masked from the prefix on; and a masked value no longer swallows a closing backquote at its end, or a closing round or square bracket at the end of a URL that has no matching opener, while a backquote inside a value never cuts it short; an address after `Bearer` or `Basic` that is itself the value of a credential label (including `Authorization:`) is masked whole, keeping the scheme word, while an address with no label stays as written. |
 | `scripts/run-engine-agent-absence-projection-test.mjs` | Absent background agents: when the engine stops reporting a background agent and no final state has arrived, the row is marked absent and this package owns every decision about it, so all clients agree. One predicate says whether a row is absent (the mark, not the status, decides). An absent row keeps its last known status, never counts as running, and is never counted as completed, failed or stopped; its elapsed time stops at the last moment it was seen, and its sentence says it may still be running. The end-of-turn sweep never settles an absent row (or a resident one). A row that comes back, or a real final state for the current cycle, clears the mark; a late final state from an earlier cycle does not. Absent rows are never removed at the short grace window. After the hard limit (30 minutes from the last time they were seen) the host is asked for the background-agent registry reading of each row: only a reading that the agent has ended or is not listed lets the row go, and each removal is returned as a fact the host must act on and announce; a reading of running, unknown, missing or unrecognised keeps the row and schedules nothing, so no standing poll is created. Until a registry reading is available every absent row stays. A row someone is viewing is held and reported separately only once the registry confirms it is gone. The row sentence, the removal sentence and the late-result sentence come from one place, never state an outcome or that the agent finished, and escape control characters in names, in the engine's removal word and in the late-result status. An end-to-end cell drives the real fleet projection and the real absence channel through every decision. From 0.84.0 the package also reads the registry itself: a reader lists the session's background registry only when the server advertises the listing, classifies its failures as unavailable (a failed read keeps the HTTP status and error code when the error carries them, and never invents either), and refuses a partly readable listing as a whole; a classifier turns the listing into the per-row reading, treating a registry status it cannot place as unknown and reading "not listed" only for keys shaped like registry handles and only when the host states the engine is a single process and names the row's session, because a key from another identity space proves nothing by being absent and, on a multi-replica deployment, a listing answered by another replica does not contain this session's agents at all (without that statement a missing row reads unknown and the row stays); and a predicate returns the agents the registry still counts as live that the host has no row for. A listing at the server's 500-row cap cannot show that an agent is gone either, so a missing row there reads unknown; the cap is checked against the engine's own list clamp and, when a server build is supplied, against the server's route. Each listing carries its session and a per-client sequence number, and a listing superseded by a later delivered read, or read for another session than the one the host names, counts for nothing. Only the exact listing object the reader returned can authorize removing a row or adding one: a spread copy, a structured clone, a filtered or a hand-built listing reads unknown and adds nothing, the listing, its row array and every row are frozen, the cap check uses the row count recorded when the page was read, and rewriting a listing's sequence number fools nothing. The fill predicate holds back agents first seen within the absence settle window — timed on one monotonic local clock from when this client's reader first saw the id, so a skew between the server's clock and this one, or reconnecting to a long-running agent, cannot skip the window — agents without a readable registration time and agents the host saw end since the read went out; the registry key of a row, progress or absence event is whichever of its two ids is shaped like a registry handle; and none of the reader, the classifier or the predicate throws. |
-| `scripts/run-plan-review-dismissal-test.mjs` | An automatic reopen does not put back a plan-review card the user closed (the first-presentation path neither checks nor clears that record, so a replayed park frame still presents its card). A plan-review card the user dismissed (Esc, abort, or any answer that is not approve or reject) is recorded per session and run at the moment of dismissal, synchronously, before anything queued behind the card can be released; a reopen marked `trigger: 'automatic'` then refuses with `{ reopened: false, dismissedByUser: true }` instead of minting a new card the user's next keystroke would land on, while the user's own next action (`trigger: 'user'`) reopens it and clears the record. The record is keyed by gate instance when the host supplies an instance reader: a new plan gate on the same run is still surfaced, and anything that cannot prove the gate is new (no reader, a failed, empty, thrown or timed-out read) refuses on the conservative side. The asynchronous form re-checks after its reads and before minting — a decision handed over meanwhile (seen by the package, or reported by the host's optional hand-over predicate) answers as "your answer is on its way"; a record that changed meanwhile makes the stale evaluation mint nothing and answer from the current record: another close refuses as the user's close and keeps the newer record, a card already back on screen (the user's own action or a concurrent automatic reopen, waited for within the receipt window and re-read once the wait is over) answers `reopened: true`, and a record that is gone (session change, ledger overflow) answers a plain refusal without `dismissedByUser`; a hand-over predicate that throws refuses with a plain `{ reopened: false }`. Each run has at most one reopen on its way: a reopen that arrives while an earlier one's card is published but not yet settled joins it instead of minting a second card and retiring the first card's answer path. Instance readers are snapshotted when they resolve, so a host that hands over its own live set still gets a new gate recognised; and a decisive-looking host answer note does not clear the record for the very card the package's own responder already judged non-decisive (the label was not on that card). A successful reopen replaces only the record taken before the card was minted (with an on-screen marker, not a deletion), a decisive answer clears it, the per-session ledger is bounded, and a session change clears its bucket. Without `trigger` the reopen answers exactly as before, apart from joining a reopen already on its way. A reopen for a session with no view subscribed to question frames (`onQuestionFrameFor(sessionKey, handler)`; `onQuestionFrame` for the default session) refuses with `{ reopened: false, _sema_noPresentationSurface: true }` — in the synchronous and the receipt form alike, per session key, and also when the view goes away while the asynchronous check is reading — and that answer comes before the user-closed check, at the entry and after the asynchronous read alike, whether or not the read proved a newer gate (after the read, a decision handed over while the read was in progress, and a dismissal record that was replaced or cleared meanwhile, still take precedence over the missing view); a host log or probe port that throws changes none of the reopen answers (the suite runs the same arrangement with a quiet and with a throwing log port and requires the same answer and the same card frames); every other refusal (receipt window lapsed, decision in flight, user-closed card, non-default slot without a delivery path, record gone) keeps its own keys without the flag. The first-presentation call does not check for a view: without one it still answers `true` and the card frame goes nowhere, which the suite records as today's behaviour, not as a guarantee. |
+| `scripts/run-plan-review-dismissal-test.mjs` | An automatic reopen does not put back a plan-review card the user closed (the first-presentation path neither checks nor clears that record, so a replayed park frame still presents its card). A plan-review card the user dismissed (Esc, abort, or any answer that is not approve or reject) is recorded per session and run at the moment of dismissal, synchronously, before anything queued behind the card can be released; a reopen marked `trigger: 'automatic'` then refuses with `{ reopened: false, dismissedByUser: true }` instead of minting a new card the user's next keystroke would land on, while the user's own next action (`trigger: 'user'`) reopens it and clears the record. The record is keyed by gate instance when the host supplies an instance reader: a new plan gate on the same run is still surfaced, and anything that cannot prove the gate is new (no reader, a failed, empty, thrown or timed-out read) refuses on the conservative side. The asynchronous form re-checks after its reads and before minting — a decision handed over meanwhile (seen by the package, or reported by the host's optional hand-over predicate) answers as "your answer is on its way"; a record that changed meanwhile makes the stale evaluation mint nothing and answer from the current record: another close refuses as the user's close and keeps the newer record, a card already back on screen (the user's own action or a concurrent automatic reopen, waited for within the receipt window and re-read once the wait is over) answers `reopened: true`, and a record that is gone (session change, ledger overflow) answers a plain refusal without `dismissedByUser`; a hand-over predicate that throws refuses with a plain `{ reopened: false }`. Each run has at most one reopen on its way: a reopen that arrives while an earlier one's card is published but not yet settled joins it instead of minting a second card and retiring the first card's answer path. Instance readers are snapshotted when they resolve, so a host that hands over its own live set still gets a new gate recognised; and a decisive-looking host answer note does not clear the record for the very card the package's own responder already judged non-decisive (the label was not on that card). A successful reopen replaces only the record taken before the card was minted (with an on-screen marker, not a deletion), a decisive answer clears it, the per-session ledger is bounded, and a session change clears its bucket. Without `trigger` the reopen answers exactly as before, apart from joining a reopen already on its way. A reopen for a session with no view subscribed to question frames (`onQuestionFrameFor(sessionKey, handler)`; `onQuestionFrame` for the default session) refuses with `{ reopened: false, _sema_noPresentationSurface: true }` — in the synchronous and the receipt form alike, per session key, and also when the view goes away while the asynchronous check is reading — and that answer comes before the user-closed check, at the entry and after the asynchronous read alike, whether or not the read proved a newer gate (after the read, a decision handed over while the read was in progress, and a dismissal record that was replaced or cleared meanwhile, still take precedence over the missing view); a host log or probe port that throws changes none of the reopen answers (the suite runs the same arrangement with a quiet and with a throwing log port and requires the same answer and the same card frames); every other refusal (receipt window lapsed, decision in flight, user-closed card, non-default slot without a delivery path, record gone) keeps its own keys without the flag. The first-presentation call does not check for a view: without one it still answers `true` and the card frame goes nowhere, which the suite records as today's behaviour, not as a guarantee. A free-text answer on the plan card (no option picked, a non-blank note) counts as a decisive answer: it clears the record and delivers exactly one rejection; a blank note, or a contradictory multi-select that carries a note, does not. |
 | `scripts/run-gate-interrupt-safety-test.mjs` | The two safety properties of running the suites themselves. The collecting runner no longer kills a suite outright when its time limit expires: it sends a termination signal first, waits a grace period for the suite to clean up, and only then kills it — and it records the suite as timed out however it exits, so a suite that exits cleanly after the signal is still not counted as passing. The limit and the grace period can be widened for a single suite in `gates-manifest.json` (an optional entry with a written reason; a malformed entry, including one with a misspelled field, stops the runner before any suite starts instead of silently falling back to the default, and an entry filed under a misspelled name is reported by this guard rather than ignored), and the negative-control suite is widened there. Every property is exercised on byte-for-byte copies of the real runner and the real negative-control suite in a throwaway directory: a suite that honours the signal finishes within the grace period, one that ignores it is killed when the grace period ends, and the summary lines are unchanged; once a suite has exited the runner waits at most a short drain window for its output pipes, so a child process that inherited them and outlives the suite neither turns a passing suite into a timeout nor holds the runner past the limit, the grace period and that window; the negative-control suite, interrupted in the middle of a rehearsal by any of the three signals or by the runner's own time limit, restores the file byte-for-byte, leaves no backup behind, stops the rehearsed guard together with anything it started, regenerates the build output (checked on a small project in the throwaway directory), and exits with 128 plus the signal number; a rebuild that would overrun the grace period is cut short and its compiler stopped; a backup left behind by an earlier run — next to a later target, loose in the source tree, or inside a symlinked dependency directory — makes it refuse to start without touching anything, naming the file and how to restore it. |
 | `scripts/run-system-reminder-open-tag-test.mjs` | The opening-tag locator for the engine's `<system-reminder>` envelope, for hosts that split a message into envelopes and body text themselves rather than only stripping or unwrapping it. `findSystemReminderOpenTag(text, from?)` returns the start and end of the next opening tag at or after `from` (UTF-16 indices, `end` one past the tag) or `null`, and it recognises exactly the two shapes the engine mints — a bare tag, or a tag with a single `mark` attribute whose value is 22 base64url characters — through the very same matcher the strip and unwrap entry points use, so there is no second grammar to drift. The guard pins both engine shapes to the exact index, refuses the non-engine shapes listed in the integration guide plus near relatives (and does not let them swallow a real tag that follows), refuses an opening tag truncated at the end of the text, answers `null` without throwing for a non-string text and for a `from` outside the integer range 0..length, and shows that the locator ignores block context (a tag inside a block body, an unclosed tag and a nested inner tag are all located — pairing with a close is the caller's loop, as in the strip entry point). On twenty thousand seeded random strings mixing both shapes, truncations, near relatives and nesting, the set of tags the locator finds equals the set derived purely from the observable answers of the strip and unwrap entry points, and a strip and an unwrap rebuilt on the locator agree with the real ones byte for byte. Two timing cells show a single call over many near-miss tags and a full walk with `from` moving forward both stay linear |
 | `scripts/run-detach-durable-off-verdict-test.mjs` | The verdict behind the durable-off hint, now a public entry point without the process-level gate. `isDetachDurableOff400(err)` answers whether an error is the engine's refusal of the detach-on-disconnect header on a deployment with no durable run ledger: status 400 and the engine's own refusal sentence in the message, and nothing else — in particular not the error code that refusal carries, because the engine uses the same generic precondition code for unrelated refusals on the same submit path, and treating those as this one would silently resubmit a turn without the header. Browser and desktop hosts, which never arm the terminal's detach ledger, can now ask the same question instead of keeping their own copy. The guard pins the positive case on the real sentence (with or without the code, with surrounding text, and on the error the SDK actually throws when a stubbed engine answers 400), the negative case on the engine's other refusals that share the code (verbatim, and again through the SDK), non-400 statuses, and eighteen malformed inputs that must answer `false` without throwing. It shows the verdict does not read the process ledger and that the hint still returns nothing before the header has been sent. The verdict never throws: an error object whose `status` or `message` getter throws, a proxy whose trap throws and a revoked proxy all answer `false`, each property is read at most once (and `message` not at all unless the status is 400), and the hint therefore answers nothing for those objects instead of throwing as it did before. Against a frozen copy of the previous hint, every other input gives the same answer byte for byte, armed or not. When the engine's source tree is available, it also checks that exactly one refusal site with that code carries the sentence and that every other one is answered `false` |
@@ -443,15 +443,15 @@ guard still cross-checks the table by name).
 | `scripts/run-session-policy-refused-removal-test.mjs` | The way out for a session that already carries a tool name the engine refuses. The engine refuses to start a run when any rule applied to it names a retired tool, or a name containing "__" without a protocol prefix, so a session whose own rule record holds such a name fails at startup on every run. The guard pins three pieces. First, the refusal test itself: it is compared name by name, in both directions, with an independent reading of the installed engine's own tables (retired names and protocol prefixes) over a corpus that includes padded, wildcard and parenthesised forms, and it is shown to be wider than the verdict used before a write — a padded or wildcard form is withheld there for another reason yet still refuses the run, so a removal driven by that verdict would leave the session broken. Every corpus name is then written alone into the deny list and into the allow list of a real engine run: the test says "refused" exactly when the run fails at startup with the refusal code and the model is never called, while the same shapes in the command lists are not audited at all. Second, the removal verb: it takes only the refused entries out of those two lists and leaves every other entry, list and ordering byte-for-byte, keeping an emptied allow list as a present empty list; after it runs, the same session's next real engine run gets past startup, for every refused name in the corpus and in both lists. Against the installed engine as an ordinary caller, taking a deny entry out is refused as a loosening — reported as such, written exactly once, the record unchanged and the next run still refused, never a false success — while an allow-list-only removal, which narrows, succeeds; against a model of the newer contract the engine has announced, the removal succeeds, and the same model still refuses a removal of anything else. A concurrent writer causes exactly one re-read, judged again on the other writer's record, and a second collision is reported rather than retried; read failures, coded write refusals and unconfirmed writes (no verdict, a receipt that cannot be read, or a receipt that does not account for the removal, including one that still holds a removed entry) each land in their own arm, and running the verb again after an unconfirmed write reports nothing left to remove. Third, the wording: one sentence per outcome; the loosening refusal names operators and says the record is unchanged, the unconfirmed one says the change may already be in effect, and none carries a rule string or engine text. The startup-failure sentence answers to that one code only, lists every place the entry can be set without naming who may remove it, and a real run with the name in the caller's own policy rather than in the session record yields the same code — which is why the sentence may not claim the entry is in the session record. The verb also requires the tool roster reported by a run on the deployment: a name listed there, as a tool name or an alias and compared exactly as the engine compares it, is left in place even when it is a retired name, because the engine accepts that rule — with a real engine run that mounts a deployment tool under a retired name (or with that name as an alias), the removal keeps the deny rule and the next run still starts. Without a roster, or with one that cannot be read, nothing is read or written and the outcome says why; a readable empty roster is still a roster. The removal sentence speaks separately about names taken out of the deny list and out of the allow list. |
 | `scripts/run-fixture-flat-done-ratchet-test.mjs` | Test fixtures that feed the engine's terminal `done` frame are kept in the shape the live engine actually sends. Since the terminal record became a tagged cause (`result.terminal`), the older flat shape (`result.status` plus loose keys) reaches a client only in two ways: a stored row replayed verbatim from before the upgrade, and the server's own refusal envelope — so a fixture written in the flat shape tests the replay path while claiming to test the live one. The suite finds every `{ type: 'done', result }` literal under `scripts/`, follows `result` to the object literal it really comes from (in place, through a variable, through a helper's parameter at each of its call sites, through a spread, or through the rows of an iterated array) and counts the flat ones that carry no one-line note saying they model a replayed row or a refusal envelope. That count may only go down: it is a ceiling kept in the ratchet registry, and a count under the ceiling prints a step-down line instead of passing in silence. A planted corpus with a known number of flat, noted and cause-shaped frames in each of those forms must be counted exactly, and a note that sits inside a string or gives no reason does not count |
 | `scripts/run-authority-envelope-mirror-test.mjs` | The authority-envelope tags that a peer's message body is defused against before it is written into a transcript line. Tags the engine uses to speak with its own authority (reminders, completion notices and the like) must never survive inside a body a peer wrote, or a forged completion notice could be read back on resume as a real one and poison the dedup ledger. The package's list is generated at build time from the engine package's public entry point, and the guard checks that the source binds that generated list rather than a hand-written copy and that it matches the installed engine package in both directions: a tag the engine treats as authority and the package lacks is red, and so is a tag the package defuses that the engine does not, since that rewrites ordinary text in a peer's message. It also confirms the published list is the engine module's own value and is still derived from the engine's envelope registry, which it reads from the engine's module by path because the registry itself is not published. At the rendered output, every engine authority tag placed in a body (opening, closing, with attributes, upper case) comes out defused on all three peer lanes, and the engine's non-authority envelope tags come out untouched |
-| `scripts/run-upstream-tables-test.mjs` | Engine facts the package may not import at run time (the notice code register and its audience table, the reasons an injected MCP server is dropped, the protocol namespace prefixes and the retired tool names with their current names) are generated at build time from the installed engine package's own public entry point into plain literal modules, which are committed. The generated files must be byte-identical to what the installed engine produces today, no other file may sit beside them, each names the generator and the engine version it came from, none imports anything, and the package's public names are the generated objects themselves. The generator refuses rather than guesses: a table that is missing or empty, a notice code without an audience row (or the reverse), a third audience value, or a retirement note in a shape it does not know all stop generation instead of producing an empty or partial table, and it reads only the engine package installed in this package's own dependency folder. Two word lists the SDK already exports as values (who asked a question, and which word a mandated approval stands on) are now the SDK's own arrays rather than copies; because those arrays are not frozen, the package's own decisions are shown not to change when they are modified in place. Inside the package, the thinking-level and permission-mode vocabularies, the tier words, the rejection sentence and the cancelled code each exist exactly once in the source. A second batch covers engine fact tables that newer engine packages publish from their entry point (the notice cause and source lists, the run-ending reasons, the reserved control-verb names, the workflow size caps, the authority envelope tags, the denial layers, the reasons a requested permission mode did not take effect, and the structured tool-card types): each is generated the same way and bound where it is used, no hand-written copy of any generated table may remain in the source, and an engine package too old to publish them makes the generator stop and name every missing export rather than fall back to reading private files. A third batch covers the closed set of codes the engine refuses a declared caller tool with: it is generated the same way and folded into the package's configuration-refusal recognition set, which must contain every one of those codes; the set only recognises them (no code path branches on them), an unexpected code shape stops generation, and an engine package that does not yet publish the set makes the generator stop and name it. A fourth batch (from 0.88.0) covers the closed set of reasons a permission mode's release of a question may carry: it is generated the same way, the package's public word list is that generated object, no hand-written copy of its words remains in the source, and an engine package that does not yet publish it makes the generator stop and name it. A sixth batch (from 0.90.0) covers the table of tool-name aliases that hook matchers resolve (the alias spelling of a built-in tool to its canonical name): the engine publishes no ready-made table, so the generator expands the aliases declared on each row of the engine's static tool catalog, and the package reads the result through two public functions (the table itself is not exported). Those functions must agree name by name with the engine's own name resolution, leave every name outside the table untouched (empty strings and inherited object keys included), never fold case or whitespace, and never throw. The generator stops rather than writing an empty table when the catalog declares no alias at all, when a catalog row has an unexpected shape, when an alias collides with another row's canonical name, or when an alias is an integer-like string (object keys of that shape would be reordered), and a run of it against a catalog without aliases exits non-zero and leaves every generated file byte-identical; the generated table is built from key-value pairs and is executed in the check against the engine's own resolver, so an alias spelled like an inherited object key still lands as an own entry The generator also produces five closed sets that the engine exports from its package root — the reasons a rule's evidence can be absent in an approval question, the lanes of the system injection queue, the full list of built-in read-deny tiers, the tiers active by default, and the per-class arbitration switches — as frozen literal tables that the package's own root entry hands out as the very same objects. The check compares them item by item and in order with the engine (the arbitration rows bit by bit, with the key set compared to the engine's own list of ask classes, and the tier lists with the engine's own tier resolver), refuses to generate when a name is missing, a list is empty or has duplicates or non-strings, a default tier is not in the full list, or an arbitration row has a missing, extra or non-boolean switch, and a run of the generator against an engine without one of them exits non-zero and leaves every generated file byte-identical. The reader that answers whether a standing answer may settle an ask reads only the generated arbitration table: its answers follow a table with one switch flipped, a class added or a class removed, the source holds no class name literal, and it treats a throwing getter, a prototype-only field or an unknown class as unknown rather than as permission. The engine code string for a self-contradictory request body appears as a string literal exactly once in the source; the three refusal readers that use it keep their own criteria. From 0.92.0 two more batches come from the engine's package root — four closed sets that stay inside the build, and nine facts the package used to copy by hand (settlement words, classifier refusal causes, the auto-mode unavailable causes, tier order and the familiar tier aliases, the background agent handle shape, the sender permission classes, MCP liveness words and the three rule behaviors) — and a third source is read: the settings contract package, from which the engine-owned hook events, the hook events whose matcher reads a tool name, the known-secret literal shape, the tier group name shape, the budget keys grouped by value type and the model entry interface words are generated into their own file. Each former copy gets a check that the public name is the generated object itself, or, for module-private constants, that the initialiser is the imported generated identifier; regex and alias-table copies are scanned for alongside array copies; and malformed upstream shapes, extra members and missing names each make the generator refuse. Two numeric limits are generated as well (the largest inline image in base64 characters, and the longest follow-up text): the same value as the engine's export, never zero, absent or malformed, not on the root entry, read by the human-submission assembly through the generated constants with no hand-written copy left there. The closed set of causes by which a deployment hook can keep a prompt from being submitted is generated as well: the same words in the same order as the engine's export, frozen, a literal tuple type, exported at the root as the generated object itself, agreed with by the engine's own membership test, and with no copy of its words in the reader that uses it; an extra member changes the generated output, while a malformed shape or a missing name makes the generator refuse. Since 0.94.0 the generator also derives the request-side permission-mode words and the user-input lane table (lane to notice code and count key); the package-side fold of the arbitration table takes an optional question face, reads the releasable classes per face from the generated ask-face tables only, and is diffed against the engine's own fold on every face. From 0.94.0 the settings-contract batch also generates the private-key header definition, which must be unanchored and judges the same samples as the contract's own check; a missing, anchored or stateful definition makes the generator refuse, and the display module's header constant is the generated object itself. Since 0.94.0 it also generates the ask-face tables (the face words, the tool names on the shell face derived from the engine's own face function over its catalog, and the releasable classes per face), checks them against the engine item by item and against the arbitration table, and refuses to generate on every malformed shape or missing export. |
-| `scripts/run-retirement-ledger-test.mjs` | Transitional code whose retirement condition used to live only in a comment now has a machine-readable trigger. `scripts/retirement-ledger.json` lists each piece (a compatibility read, a mirrored table, a normalizer inside the differential check) with an anchor into its own code, the retirement condition quoted from the source comment, and one of three trigger kinds: this package's version reaching a retire-by version; the installed upstream package reaching a version, exporting a name from its entry, or declaring a member on a given type (read with the TypeScript checker); this package's declared support floor (`scripts/support-floor.json`) for the server, the engine or the SDK reaching a given version (an unreadable or malformed declaration fails the check instead of being skipped); or the reference client tree used by the differential check having moved to a given version of this package. The check fails when a trigger has fired and the transitional code is still there, when a row's anchor can no longer be found (the code is gone but the row stayed), when the quoted condition no longer matches the source comment, when a row has no machine trigger at all, when a retire-by version sits more than three minor lines ahead, and when the anchor of an already retired piece reappears. A reading that cannot be taken is reported as skipped when the absence is legitimate (no reference tree on disk, an upstream declared only as a peer and not installed) and is a harness fault otherwise. Before giving any verdict the judge proves itself on in-memory fixtures and on one known-present and one known-absent name in the installed upstream types. Rows already moved to the retired list may also carry the names, keys and function arities that the retirement removed; for those rows the check confirms the names are absent from the package root entry (as exports and as runtime keys), from the built output, and from every check script's code positions (comments excluded), that each retired function has its reduced parameter count (taken from every type signature as well as the runtime length, so an added optional, default or rest parameter is seen), and that a compile-time pin on the root entry's keys holds in both directions — no retired name present, every known live name still present, with a planted live name that must fail. |
+| `scripts/run-upstream-tables-test.mjs` | Engine facts the package may not import at run time (the notice code register and its audience table, the reasons an injected MCP server is dropped, the protocol namespace prefixes and the retired tool names with their current names) are generated at build time from the installed engine package's own public entry point into plain literal modules, which are committed. The generated files must be byte-identical to what the installed engine produces today, no other file may sit beside them, each names the generator and the engine version it came from, none imports anything, and the package's public names are the generated objects themselves. The generator refuses rather than guesses: a table that is missing or empty, a notice code without an audience row (or the reverse), a third audience value, or a retirement note in a shape it does not know all stop generation instead of producing an empty or partial table, and it reads only the engine package installed in this package's own dependency folder. Two word lists the SDK already exports as values (who asked a question, and which word a mandated approval stands on) are now the SDK's own arrays rather than copies; because those arrays are not frozen, the package's own decisions are shown not to change when they are modified in place. Inside the package, the thinking-level and permission-mode vocabularies, the tier words, the rejection sentence and the cancelled code each exist exactly once in the source. A second batch covers engine fact tables that newer engine packages publish from their entry point (the notice cause and source lists, the run-ending reasons, the reserved control-verb names, the workflow size caps, the authority envelope tags, the denial layers, the reasons a requested permission mode did not take effect, and the structured tool-card types): each is generated the same way and bound where it is used, no hand-written copy of any generated table may remain in the source, and an engine package too old to publish them makes the generator stop and name every missing export rather than fall back to reading private files. A third batch covers the closed set of codes the engine refuses a declared caller tool with: it is generated the same way and folded into the package's configuration-refusal recognition set, which must contain every one of those codes; the set only recognises them (no code path branches on them), an unexpected code shape stops generation, and an engine package that does not yet publish the set makes the generator stop and name it. A fourth batch (from 0.88.0) covers the closed set of reasons a permission mode's release of a question may carry: it is generated the same way, the package's public word list is that generated object, no hand-written copy of its words remains in the source, and an engine package that does not yet publish it makes the generator stop and name it. A sixth batch (from 0.90.0) covers the table of tool-name aliases that hook matchers resolve (the alias spelling of a built-in tool to its canonical name): the engine publishes no ready-made table, so the generator expands the aliases declared on each row of the engine's static tool catalog, and the package reads the result through two public functions (the table itself is not exported). Those functions must agree name by name with the engine's own name resolution, leave every name outside the table untouched (empty strings and inherited object keys included), never fold case or whitespace, and never throw. The generator stops rather than writing an empty table when the catalog declares no alias at all, when a catalog row has an unexpected shape, when an alias collides with another row's canonical name, or when an alias is an integer-like string (object keys of that shape would be reordered), and a run of it against a catalog without aliases exits non-zero and leaves every generated file byte-identical; the generated table is built from key-value pairs and is executed in the check against the engine's own resolver, so an alias spelled like an inherited object key still lands as an own entry The generator also produces five closed sets that the engine exports from its package root — the reasons a rule's evidence can be absent in an approval question, the lanes of the system injection queue, the full list of built-in read-deny tiers, the tiers active by default, and the per-class arbitration switches — as frozen literal tables that the package's own root entry hands out as the very same objects. The check compares them item by item and in order with the engine (the arbitration rows bit by bit, with the key set compared to the engine's own list of ask classes, and the tier lists with the engine's own tier resolver), refuses to generate when a name is missing, a list is empty or has duplicates or non-strings, a default tier is not in the full list, or an arbitration row has a missing, extra or non-boolean switch, and a run of the generator against an engine without one of them exits non-zero and leaves every generated file byte-identical. The reader that answers whether a standing answer may settle an ask reads only the generated arbitration table: its answers follow a table with one switch flipped, a class added or a class removed, the source holds no class name literal, and it treats a throwing getter, a prototype-only field or an unknown class as unknown rather than as permission. The engine code string for a self-contradictory request body appears as a string literal exactly once in the source; the three refusal readers that use it keep their own criteria. From 0.92.0 two more batches come from the engine's package root — four closed sets that stay inside the build, and nine facts the package used to copy by hand (settlement words, classifier refusal causes, the auto-mode unavailable causes, tier order and the familiar tier aliases, the background agent handle shape, the sender permission classes, MCP liveness words and the three rule behaviors) — and a third source is read: the settings contract package, from which the engine-owned hook events, the hook events whose matcher reads a tool name, the known-secret literal shape, the tier group name shape, the budget keys grouped by value type and the model entry interface words are generated into their own file. Each former copy gets a check that the public name is the generated object itself, or, for module-private constants, that the initialiser is the imported generated identifier; regex and alias-table copies are scanned for alongside array copies; and malformed upstream shapes, extra members and missing names each make the generator refuse. Two numeric limits are generated as well (the largest inline image in base64 characters, and the longest follow-up text): the same value as the engine's export, never zero, absent or malformed, not on the root entry, read by the human-submission assembly through the generated constants with no hand-written copy left there. The closed set of causes by which a deployment hook can keep a prompt from being submitted is generated as well: the same words in the same order as the engine's export, frozen, a literal tuple type, exported at the root as the generated object itself, agreed with by the engine's own membership test, and with no copy of its words in the reader that uses it; an extra member changes the generated output, while a malformed shape or a missing name makes the generator refuse. Since 0.94.0 the generator also derives the request-side permission-mode words and the user-input lane table (lane to notice code and count key); the package-side fold of the arbitration table takes an optional question face, reads the releasable classes per face from the generated ask-face tables only, and is diffed against the engine's own fold on every face. From 0.94.0 the settings-contract batch also generates the private-key header definition, which must be unanchored and judges the same samples as the contract's own check; a missing, anchored or stateful definition makes the generator refuse, and the display module's header constant is the generated object itself. Since 0.94.0 it also generates the ask-face tables (the face words, the tool names on the shell face derived from the engine's own face function over its catalog, and the releasable classes per face), checks them against the engine item by item and against the arbitration table, and refuses to generate on every malformed shape or missing export. From 0.95.0 a settings-contract function batch copies three pure functions from the contract package's entry point (the narrow loopback check used before sending no credentials, the hierarchical URL check, and the @-mention narrowing) by their own source text: the generator refuses when a function reaches for any name outside itself except an allowed global, the copies must answer exactly like the originals on independent samples, and the package's readers use the copies instead of their own versions. The configuration-refusal registry has no read point inside the package (the only reading of "is this a configuration refusal" is the prefix predicate), and its own comment says so. |
+| `scripts/run-retirement-ledger-test.mjs` | Transitional code whose retirement condition used to live only in a comment now has a machine-readable trigger. `scripts/retirement-ledger.json` lists each piece (a compatibility read, a mirrored table, a normalizer inside the differential check) with an anchor into its own code, the retirement condition quoted from the source comment, and one of three trigger kinds: this package's version reaching a retire-by version; the installed upstream package reaching a version, exporting a name from its entry, or declaring a member on a given type (read with the TypeScript checker); this package's declared support floor (`scripts/support-floor.json`) for the server, the engine or the SDK reaching a given version (an unreadable or malformed declaration fails the check instead of being skipped); or the reference client tree used by the differential check having moved to a given version of this package. The check fails when a trigger has fired and the transitional code is still there, when a row's anchor can no longer be found (the code is gone but the row stayed), when the quoted condition no longer matches the source comment, when a row has no machine trigger at all, when a retire-by version sits more than three minor lines ahead, and when the anchor of an already retired piece reappears. A reading that cannot be taken is reported as skipped when the absence is legitimate (no reference tree on disk, an upstream declared only as a peer and not installed) and is a harness fault otherwise. Before giving any verdict the judge proves itself on in-memory fixtures and on one known-present and one known-absent name in the installed upstream types. Rows already moved to the retired list may also carry the names, keys and function arities that the retirement removed; for those rows the check confirms the names are absent from the package root entry (as exports and as runtime keys), from the built output, and from every check script's code positions (comments excluded), that each retired function has its reduced parameter count (taken from every type signature as well as the runtime length, so an added optional, default or rest parameter is seen), and that a compile-time pin on the root entry's keys holds in both directions — no retired name present, every known live name still present, with a planted live name that must fail. Rows may carry a structured blockedOn (upstream ticket + landing criterion + a ruling id cited in the ruling text): a due row with it prints as BLOCKED-DUE instead of failing, capped at 3 such rows. |
 | `scripts/run-plan-review-decision-ledger-test.mjs` | The plan-review reopen path refuses on one decision ledger shared by every delivery path, not only the package's own POST window: a decision the host reports as handed over (before the card's answer reaches the package, or on the host's own delivery path) counts as in flight until it settles; a pending-row read that started before a decision settled is a stale snapshot of the gate just answered, so a reopen carrying that read mark refuses; and a decision proven not to have taken effect (nothing sent, a refusal that by contract changed nothing, or the engine proven to still hold the very same gate instance — a status alone does not prove it, since a run that moved on to a new plan gate reports the same status) neither counts as in flight nor makes such a snapshot stale, so a failed decision does not consume a legitimate reopen. The ledger's capacity bound never evicts the package's own in-flight decisions, so a duplicate approval stays latched however many decisions are in flight. The package's own in-flight window now runs from the POST to the moment the outcome is emitted (settlement happens before the outcome is queued, so a read the host starts from the queue port is not stale), while the duplicate-delivery latch keeps its POST-only window. A handed-over decision the package's delivery path then sends is the same ledger entry (the host's early settle does not release it); two decisions on one run each settle on their own; positive evidence that the run left the gate settles what is open without releasing the latch; bad inputs neither throw nor record. The ledger is checked against the terminal's current implementation vector by vector when that tree is on disk. Positive evidence that the run left the gate can arrive after the host has handed a decision over and before the package's delivery path picks it up (while the host is still waiting on its gate snapshot, or in the microtasks between that wait ending and the delivery call); the delivery path then still picks up that same entry instead of minting a new one, so the latest decision mark the host keyed its snapshot and its not-applied notice on does not move, the duplicate-delivery latch and the capacity rule hold for it, and the three failure outcomes read byte for byte as they would without the hand-over. Only an entry settled by such evidence is picked up again: one the host settled itself, one it revoked, and one that already produced an outcome are not, and a new decision gets a new mark |
 | `scripts/run-subagent-injected-wire-test.mjs` | The background-agent verbs — the child report read, the live tail, the task-handle output read and stop, steer, the resume assembly, manual compaction with its capability warm-up, and the delegated-prompt read — accept a host-supplied engine client, and the plan-review orchestration now resolves its client through the same single construction point, so a browser host behind a same-origin relay runs every one of these paths without an installed engine target. With a client injected, each entry sends only through it: proven against a real relay-form SDK client and a fake engine while the installed target (default and keyed slots alike) points at a different fake engine that must receive nothing, with the row's own session parameter carried, no `authorization` header on any request (checked beside a token-bearing client on the same spy), and a planted token never reaching a log line, a return value or a failure detail on the injected, installed-target and unreachable paths. The capability evidence behind the task-handle verbs and the row stop gate is read only under the key the host names, never from the installed target; the report, tail and compaction warm-up cache their own probe under that key and re-probe when it is absent. A malformed injected client or connection lands on each entry's existing cannot-send outcome (`null`, `no-wire`, `unavailable`, `offline`, `unreadable`) with zero requests, never a throw or an unhandled rejection. Every request through an injected client settles at the package's own deadline even while the client sleeps through a retry back-off and aborts the request underneath; the caller's own signal still applies, and the old signatures answer as before apart from the no-answer outcomes described next. A steer, resume or stop that may have reached the engine but got no answer — the package's own deadline on an injected client, the client's per-request limit on either path, or a transport failure that cannot be shown to have happened before anything was sent (the connection dropped after the request went out) — is reported with a machine-readable `unconfirmed: true` beside the unchanged `reason`, so a host can say it may have taken effect instead of calling it a failure; a cancellation by the caller after the request went out is reported the same way (only the wait was cancelled), while a real 4xx or 5xx answer, a connection that could not be made at all, a verb that threw before returning a promise, and a cancellation already in place before the call (nothing is sent) carry no such key. An installed target is copied by value when a call resolves it, so a delayed or retried compaction still goes to the engine and credentials it was resolved against even if the host's target object changes meanwhile, and a verb that throws synchronously is classified exactly like one that rejects. The resume failure classifier that hosts reuse for their own direct calls makes the same judgement from the same single source, so every client renders it the same way. A successful sub-agent steer carries the server's acceptance id (`messageId`) only when the receipt holds a non-empty string; older servers leave it out. |
 | `scripts/run-session-background-stop-test.mjs` | The session-level stop for background work (`stopEngineSessionBackground`): one call asks the engine to stop the background tasks of a session, with `includeRetained` passed through as the required true / false choice. It resolves its client through the same single construction point as the other engine verbs (a host-supplied client, or the installed engine target for the chosen session slot) and sends nothing unless the engine has declared the stop face (`background.exitFaces` strictly `true`) under the capability key that applies: evidence that is missing or unread, an older engine without the face, a malformed flag, a flag found only on the prototype chain, or evidence recorded for a different deployment all end in `unavailable` with zero requests. The request body on the wire is exactly `{"includeRetained": <boolean>}` whatever object the caller passes. A 200 answer returns the receipts verbatim (identifier and outcome word; a row whose word it does not recognise is kept with that word as it is, so the host can still name it and the companion below sorts it as may still be running), counts rows whose identifier or word it cannot read instead of inventing them, and leaves the receipt list absent rather than empty when the body cannot be read. Refusals are sorted by machine code before HTTP status (missing service credential, no ownership face, unknown session versus missing route on the same 404, request shape, not permitted, rate limited with the server's wait hint, anything else), a server failure is an `error`, and each call sends exactly one request even through a client configured to retry. A request that may have reached the engine but got no answer — a time limit, a connection that dropped after the request went out, or a cancellation by the caller after the request went out — carries `unconfirmed: true`, while a real answer, a connection that never opened, or a cancellation already in place before the call (nothing is sent) does not. Malformed arguments, options, connections, clients, capability records, answer bodies and error objects never throw and never leave an unhandled rejection, and a planted token never reaches a log line or a result. A pure companion (`engineSessionBackgroundReceiptReadingOf`, with the frozen state list `ENGINE_SESSION_BACKGROUND_RECEIPT_STATES` derived from its wording table) sorts each receipt word into stopped, still running (kept on purpose) or may still be running, with one fixed sentence per state so every client says the same thing: each of the four words lands in its state, while any word it does not recognise — a future word, a different letter case, stray whitespace, a near spelling, a prototype-chain name — lands in may still be running with the word carried back verbatim, and non-string input (boxed strings and objects with their own `toString` included) lands there too without being coerced and without throwing; it never reports stopped for anything it cannot confirm. The three sentences differ, the two that are not stopped never use the word stop, and the state list and the states actually produced match in both directions. |
 | `scripts/run-sealed-key-capability-test.mjs` | The engine's sealed-key discovery segment (`capabilities.sealedKey`), read the same four-state way as the other capability readers: exactly three non-empty strings (`alg`, `publicKeyId`, `publicKey`) make a usable reading and are passed through as they are; an absent key means *do not seal* (an older engine, or an engine that could not set up its key custody — the same action either way) and is never treated as a key; a present but malformed segment is dropped rather than turned into a half reading; and anything the segment carries beyond the three public fields — a private key, a creation time, anything else — never reaches the reading, so this reader cannot become a credential channel. The reading is checked against the projection a `7.104` engine package actually mints, both with a key store and without one. |
 | `scripts/run-session-integrity-refusal-test.mjs` | Three refusals a session can meet on newer servers, read the same way on every client. When a session's saved history cannot be read, the server answers with one code whether the refusal arrives as an HTTP error, as the failure recorded at the end of a streamed run, or on the background run's record; one predicate decides it for all three carriers, it keys on the code alone — never on the status or on the engine's wording — and one sentence explains it: the history is damaged, retrying will not help, whoever operates the engine has to repair it, and meanwhile a new session works. The package's own end-of-run error line and the error of the result frame both carry that sentence, identically, and nothing changes for any other code. When a delete is refused because background agents the session started are still running somewhere this server cannot stop them, the refusal is read with the ids it names — each id checked on its own, a bad entry dropped rather than invented, and a list with no usable entry reported as absent rather than as an empty list, which would read as "no agents left"; a refusal because a run is still in progress is read with that run's id, and any other conflict falls back to a sentence that does not guess. When a decision reaches an approval that moved in the meantime (stopped, decided elsewhere, parked again, claimed by another decision, or — newly — belonging to a session that was deleted), the package reads the code rather than the error class, says the decision was not applied and that the pending approvals should be fetched again, and the approval legs treat it as "the approval is no longer where it was" and keep reading the run, instead of letting the wording of the server's sentence decide. When the server package is supplied, the guard drives the server's own code for each of these answers and pushes the result through the SDK's error mapping before the package reads it. |
 | `scripts/run-mode-release-projection-test.mjs` | The permission-mode fields on a tool call's gate record. Newer engines record, on an allowed call, that a permission mode answered a question nobody was asked (which mode, which question class, the whole class set, and the origin and mandate words the card would have shown), and on a refused call that the session's permission mode itself refused it (which mode and which question class). The package's gate reader carries both into its view — the allowed-call account whole or not at all, the refusal fields one by one, unfamiliar mode or class words passed through as written — and exposes a named reader for the allowed-call account. Absence is not a negative: an older engine, or a server that does not forward these fields, produces a record without them, so the reader answers `undefined` rather than claiming no mode was involved; when a server package is supplied, the guard runs the same records through that server's own projection to show exactly that. The fixtures are first passed through the engine's own record screen, the view's fields are checked against the engine's declared shapes in both directions, the fields survive the projection onto the transcript arm unchanged, and a mode refusal is filed in the transcript as a permission-rule denial. From 0.88.0 the release reason word has a closed-set mirror: the public word list is the build-time generated table taken from the engine package's entry point (the same object, frozen, equal to the engine's list word for word), the membership predicate answers exactly as the engine's own predicate on members, unknown words, case and whitespace variants, empty strings, non-strings and prototype keys, and the reader still passes an unfamiliar reason through unchanged; a release minted by the installed engine's own mode arbitration reads back field for field with its reason recognised. |
-| `scripts/run-read-credential-refusal-projection-test.mjs` | The read-credential refusal on the engine's pre-gate read routes (workflow reads and the live fleet stream), projected as a named state instead of a silent retry loop or a generic failure. On a deployment that requires a service credential, these reads now answer 401 when the client presents none it accepts, and before this change the workflow monitor kept reporting "still loading" and retried every few seconds forever, while the background view folded the refusal into "unavailable". The guard drives the real monitor through a scripted transport: a 401 with the unauthorized code, or with no code, on either the detail read or the list read, yields the new `unauthorized` state carrying the last error and — when the run had been live — the last snapshot; the loop then waits on a long back-off (the timer values it schedules are recorded) rather than the short one, opens no activity stream, and returns to live once the credential is accepted. A 403, a 407 and the two other 401 codes (a missing principal, an unverifiable signed principal) do not enter this state, and for every other status and for connection errors the outcome is compared, case by case, with the previous judgement frozen inside the guard — a deployment without a service credential never reaches the new state. The background view's fleet source gets the matching `unauthorized` health word for both 401 shapes, keeps its rows out of the view, leaves the assistant source untouched and recovers to `ok`; every other failure shape is again compared with the frozen previous judgement, and the assistant source's own 401 staying `unavailable` is pinned as a documented limit. One exported sentence serves both places: it says the engine requires a credential the client did not present, and never that there is nothing to show or that the engine is down. |
+| `scripts/run-read-credential-refusal-projection-test.mjs` | The read-credential refusal on the engine's pre-gate read routes (workflow reads and the live fleet stream), projected as a named state instead of a silent retry loop or a generic failure. On a deployment that requires a service credential, these reads now answer 401 when the client presents none it accepts, and before this change the workflow monitor kept reporting "still loading" and retried every few seconds forever, while the background view folded the refusal into "unavailable". The guard drives the real monitor through a scripted transport: a 401 with the unauthorized code, or with no code, on either the detail read or the list read, yields the new `unauthorized` state carrying the last error and — when the run had been live — the last snapshot; the loop then waits on a long back-off (the timer values it schedules are recorded) rather than the short one, opens no activity stream, and returns to live once the credential is accepted. A 403, a 407 and the two other 401 codes (a missing principal, an unverifiable signed principal) do not enter this state, and for every other status and for connection errors the outcome is compared, case by case, with the previous judgement frozen inside the guard — a deployment without a service credential never reaches the new state. The background view's fleet source gets the matching `unauthorized` health word for both 401 shapes, keeps its rows out of the view, leaves the assistant source untouched and recovers to `ok`; every other failure shape is again compared with the frozen previous judgement, and since 0.95.0 the assistant source's own 401 gets the same word from the same judge in the same order (a surface that is not mounted at all is still reported as not configured first, and a capability probe that itself answers 401 does not rewrite the verdict), with both sources answering identically for any one thrown error. The workflow monitor's status keeps the HTTP status code only for the two degrades the engine states itself (501 and 404) and null for every other failure, so that whether the credential was accepted has exactly one reading, the monitor's resolution. One exported sentence serves both places: it says the engine requires a credential the client did not present, and never that there is nothing to show or that the engine is down. |
 | `scripts/run-rules-legacy-tool-name-projection-test.mjs` | The session rule write refusal for tool names the engine would refuse at run startup, projected as a named outcome. A session rule write whose tool-name lists carry a retired tool name, or a name containing "__" without a protocol prefix, is refused before anything is stored, with a dedicated code and the offending list and name at the top of the error body. The guard pins the new code constant and its family predicate (an open prefix test shaped like the configuration family's, false for anything that is not a string of that family), then feeds the failure classifier the error exactly as two SDK generations deliver it — a plain API error on the older one, a bad-request error on the newer one — and requires both to land in the named arm by code, ahead of the status. The list and name are read only from the body slot the SDK provides, narrowed to the two tool-name lists and a non-empty name, and never from the error object's own `name`, which is its class name. An unknown member of the same family lands in the generic answered-refusal arm rather than being folded into the named one or described as a malformed body. Against a model of the refusing write gate that uses the installed engine's own tables and order, the tightening flow reports the named outcome, writes exactly once and leaves the stored record untouched — both when a host sends such a name directly and when the stored record already holds one from before the gate existed (the outcome then says so), and when this package's mirror of the retired-name table is older than the engine's. The removal verb, which deliberately keeps a name the supplied roster mounts, gets the same named outcome when the gate refuses to store that name. One sentence covers the tightening outcome and the code lookup; it shares its description of the two name classes with the startup-refusal sentence and its fix with the before-write sentence, word for word, carries no rule text, and differs from every other outcome sentence. |
 | `scripts/run-store-center-wiring-capability-test.mjs` | The two cloud-mode capability positions (engine ≥7.106.0), read the same four-state way as the other capability readers. The store posture (`capabilities.store`: declared backend, assembled backend, whether it is durable, an optional degradation reason, and the SQL engine seat) is always present on a current engine, so an absent key means an older engine and is never read as “not durable”; words are carried as reported rather than folded into known ones, the SQL seat is judged by the very same rule as the SQL-posture reader, and one malformed seat discards the whole reading. The center wiring (`capabilities.centerWiring`) is present only when the engine is connected to a configuration center: with no quota lease configured its three lease-dependent keys are legitimately absent, and an absent key is split by the same response — a current engine (it carries the store posture) is reported as not connected, an older one as not reported. Two predicates live in the package (does this replica persist; is the caller's budget lease-enforced — known words matched exactly, everything else unknown), and each reader has one table of doctor sentences in which no two states read alike, the older-engine sentence never claims “not durable” or “not connected”, the two failure postures cannot be said the wrong way round, and unrecognised words are shown escaped and bounded. Both are checked against the engine's own projections: every form they mint reads back exactly, every word in their vocabularies lands on a known arm, and the member lists the engine declares for these two positions equal the reading's own keys. |
 | `scripts/run-inherit-env-wire-test.mjs` | The request field that lets the model-driven shell inherit this machine's whole environment (`"all"`) or keep the default scrubbed environment (`"scrub"`). Both words go out exactly as given on the two user lanes, an absent value is never filled in, and any other value (including a list of variable names) is refused before anything is sent. The value reaches a request only through a helper that checks the engine's version: a server whose version cannot be read gets nothing, and the caller receives the reason plus one sentence saying what the run's shell gets instead, including that leaving the field out also clears an earlier whole-environment choice on engines that know it; from 0.92.0 a readable version always gets the field (a server below the support floor refuses the request loudly rather than being skipped quietly). An invalid-field refusal of a request that asked for `"all"` is read by code with one conditionally worded sentence, because the same code has other causes. |
@@ -459,14 +459,14 @@ guard still cross-checks the table by name).
 | `scripts/run-mandated-card-values-test.mjs` | Approval cards that newer servers mark as mandatory for four kinds of question (a deployment command policy, an operator's never-auto list, an exhausted durable budget, and supervisor routing) are built with the server's own code and fed through both card paths: the card carries the mandatory mark and the matching reason word, the shared check reports it as mandatory, no mandate word is invented, and a session-wide allow that the server declines to remember produces the single not-remembered notice. An ordinary approval-list card stays non-mandatory and remembered, and the older card shapes are shown to read the other way. This check needs an installed server package to run and reports itself as skipped otherwise. |
 | `scripts/run-park-failed-reason-projection-test.mjs` | The fail-soft true cause when a parked approval cannot be decided and the engine's own `done{suspended}` terminal is the one handed back: that terminal is still returned as-is (no second terminal), but it carries `_sema_park_failed_reason` — the same sentence, byte for byte, that the durable leg puts on its synthesized terminal — and the result frame carries it through with credentials washed the same way `errors[]` is (value only; same sentence as the durable leg after washing). The key is left off when this host has no decision surface at all (no approval card port, a port installed under another session key, or no question overlay): that sentence's way out is to decide on the card, which such a host never shows. Pinned with a policy-fold refusal on decide, both stall exits, a host-retracted card, credential samples through the full bridge on both legs, the no-decision-surface forms, and the absent forms (user interrupt, malformed values, ordinary terminals). |
 | `scripts/run-tool-history-mismatch-projection-test.mjs` | When a provider rejects every request of a session because a tool call in the conversation history no longer pairs up with its result, the terminal error row carries a machine-readable key naming the kind of mismatch and the first tool call id the provider named; a host that offers the rewind command gets the same recovery sentence CC shows (unless that host names the engine and the engine reports it cannot rewind the conversation), plus a count from the second time in a row, while the print and utility lanes, hosts without that command, and the result frame's error text stay unchanged. Only the run's own failure is read, never assistant text, and only when the provider sentence opens the provider's message: other statuses, other 400s and quoted copies do not match. The count is per session and per run, ignores replays, resets on any other outcome and never overstates when it cannot tell; the row-resolving helper that finds the prompt to go back before is checked for the target, the first-turn case and every undecidable case. The error carrier the package mints for a terminal is the same one the failure line is judged on, so handing it to the edit-and-resend reading makes that reading yield exactly when the line carries the mismatch key, across thirty-one terminal shapes including accessor and inherited fields, governance and busy endings among them; the failure line and the carrier are derived from one reading of the frame, so a field whose answer changes between reads cannot make the line and its key disagree, and an empty error sentence yields neither a line nor a carrier. The key also records whether that row offered the recovery sentence, so a host reads the fact instead of re-deciding it later, and the key reader itself is exported and checked from the package entry. The sentence naming the broken turn, which a host appends once it has resolved that turn, is checked for its two exact forms, cleaning before cutting at sixty code points, linear tag stripping, and byte equality with the shipped terminal wording. On any other provider rejection the same terminal row ends with the edit-and-resend sentence and carries that reading as a key: never together with the mismatch key, never on governance, busy or model-output rows, and every row without one of these changes is byte-identical to the previous release. The check also feeds outcomes built and published by an installed server package when one is available and reports that part as skipped otherwise. The arm of the synthesized terminal row is decided in one function that both the row text and the error carrier read, and its order is pinned: a lock or governance terminal keeps its row even when the error text starts with the model-output sentence. |
-| `scripts/run-display-untrusted-baseline-diff-test.mjs` | The credential side of the display outlet (`displayUntrusted` and `washCredentials`), diffed byte for byte against the previous published release: five input pools (seeded random strings, exhaustive shapes around labels, quoted values, userinfo, schemes and invisible units, and URL parameters separated by `;` or `,`) run through fifteen outlets. A secret the previous release masked must stay masked wherever it sits in a credential position; the outlet with only the URL net enabled is judged within that net's own reach. Clean lines from the package's own documents pass through unchanged, and every outlet is idempotent (masking the output again changes nothing). The previous release tarball is supplied through `SEMA_DU_BASELINE_TGZ`; without it the diff sections report as skipped and the run counts as partial, not green. The minimal-shape pool also carries scheme-less userinfo whose password holds a quote or angle bracket after a credential label, and the shapes that once washed differently on a second pass. A full-product grammar pool of 64,512 strings joins the idempotence check, and every pool is also checked for the credential stage reaching its pass cap (it must never do so). The pools now include the bare `token` scheme-word forms and a full `token` grammar product, so the new wash is held to the same idempotence and no-regression bars against the previous release. They also include label and `bearer` / `basic` values that run into a `token` scheme word, a non-ASCII space and an address, with or without a leading character, and quoted values where `token` and an address follow `bearer`. A YAML block-scalar pool (block headers with every chomping and indentation indicator, comments, raw and JSON-escaped line breaks, non-ASCII and tab indentation, invisible characters at line starts, credentials, control strings and bidi characters inside the block, and seven kinds of line after the block) joins all checks: no secret inside a block may show on any outlet that runs the word net, the previous release must show some on the same pool, and the non-block forms in that pool must stay byte for byte as in the previous release. A sixth pool covers lists and YAML node properties after a credential label: block, flow and comma lists plus tag and anchor prefixes, alongside look-alike non-list forms. The secrets in the list pool must stay masked on every outlet, the look-alike forms must read byte for byte as in the previous release, and the three forms whose reading changed on purpose (a lone dash after the colon, a value after a dash, a block sequence of plain key names) are pinned as registered divergences. |
+| `scripts/run-display-untrusted-baseline-diff-test.mjs` | The credential side of the display outlet (`displayUntrusted` and `washCredentials`), diffed byte for byte against the previous published release: five input pools (seeded random strings, exhaustive shapes around labels, quoted values, userinfo, schemes and invisible units, and URL parameters separated by `;` or `,`) run through fifteen outlets. A secret the previous release masked must stay masked wherever it sits in a credential position; the outlet with only the URL net enabled is judged within that net's own reach. Clean lines from the package's own documents pass through unchanged, and every outlet is idempotent (masking the output again changes nothing). The previous release tarball is supplied through `SEMA_DU_BASELINE_TGZ`; without it the diff sections report as skipped and the run counts as partial, not green. The minimal-shape pool also carries scheme-less userinfo whose password holds a quote or angle bracket after a credential label, and the shapes that once washed differently on a second pass. A full-product grammar pool of 64,512 strings joins the idempotence check, and every pool is also checked for the credential stage reaching its pass cap (it must never do so). The pools now include the bare `token` scheme-word forms and a full `token` grammar product, so the new wash is held to the same idempotence and no-regression bars against the previous release. They also include label and `bearer` / `basic` values that run into a `token` scheme word, a non-ASCII space and an address, with or without a leading character, and quoted values where `token` and an address follow `bearer`. A YAML block-scalar pool (block headers with every chomping and indentation indicator, comments, raw and JSON-escaped line breaks, non-ASCII and tab indentation, invisible characters at line starts, credentials, control strings and bidi characters inside the block, and seven kinds of line after the block) joins all checks: no secret inside a block may show on any outlet that runs the word net, the previous release must show some on the same pool, and the non-block forms in that pool must stay byte for byte as in the previous release. A sixth pool covers lists and YAML node properties after a credential label: block, flow and comma lists plus tag and anchor prefixes, alongside look-alike non-list forms. The secrets in the list pool must stay masked on every outlet, the look-alike forms must read byte for byte as in the previous release, and the three forms whose reading changed on purpose (a lone dash after the colon, a value after a dash, a block sequence of plain key names) are pinned as registered divergences. A seventh pool covers the 0.95.0 credential-net shapes (bracketed values after scheme words and labels, a JSON-escaped tab before the value, unlabelled random-looking tokens built inside the shape rule, associative-array arrows, a label followed by a `token` scheme word and an address, and known prefixes glued after an underscore or hyphen) alongside normal long strings and placeholders: the secrets must stay masked on every outlet while the previous release shows some, the normal strings must read byte for byte as in the previous release, eleven readings that changed on purpose are pinned as registered divergences, and the share of purely random tokens the shape rule catches is read at four lengths with a floor of 0.9. |
 | `scripts/run-registry-conflict-projection-test.mjs` | How a client tells the two meanings of a registry write conflict apart. The registry answers the same 409 when a write merely lost an optimistic-lock race (re-read, re-base, write once more) and when the space being written was deleted, or deleted and re-created under the same name, while the write was in flight (nothing was written, and a re-read would return the replacement space, so a retry would put an old draft into a space nobody reviewed). The package ships the same three-word classification the newer SDK uses under the same name, plus a predicate for caught errors, because the SDK version this package supports does not have them yet. The guard checks every status and body combination against the newer SDK's own function when that SDK is available, checks the predicate against the newer SDK's own error classes, reads every 409 the registry service mints from its source when that tree is available and replays the resulting bodies, and checks that unreadable bodies never throw and never classify as the retryable kind. The parts that need the newer SDK or the registry source report as skipped when those are absent. |
 | `scripts/run-prompt-delivery-projection-test.mjs` | Whether a user message actually reached the engine, judged once for every client. A ledger records, per message id and per submitted batch, that a submission entered the submit path and whether the engine created a run for it: a stream frame from the new run counts, a later accepted steer into the running run counts, while the session-busy refusal does not, and neither do the frames a server writes before creating the run (the stream-head `meta` frame and replays of approval cards still pending from earlier runs), which arrive ahead of that refusal. Both sets only grow, so the answers do not depend on arrival order, replays or repeats; a message never seen on the submit path is reported as unknown rather than undelivered. Two row-level readers filter side-channel payloads (only messages known to be undelivered are removed, everything else is kept in order). Records serialize one per line in the shape hosts already persist. The check also replays the exact frames an installed server package writes for a busy session through the real SDK stream parser when one is available and reports that part as skipped otherwise. The idempotent-replay form of the busy-session refusal (the raw conflict body without a status, sent when a host reuses an idempotency key while the first submission is still in flight) does not count as delivered either. Since 0.90.0 the check also feeds the delivery frames a steer now produces (a queued frame and an applied frame on one input id, which equals the message id on the receipt): the first records delivery, the second changes nothing, replays and reversed order give the same records, and the answer does not depend on whether the input id equals the message id, differs from it, or is absent. The ledger object itself also carries the two row-level readers as methods bound to it, `knownUndeliveredMessage(message)` and `withoutUndelivered(messages)`, which answer exactly what the two free functions answer for that ledger on every input shape: a bare string is not a message row, so it is judged false, and that is why the row-level method does not reuse the name of the id-keyed `knownUndelivered(uuid)`. They keep answering when taken off the ledger, when the ledger is copied, and after it is frozen, they read the ledger as it is now, and a ledger rebuilt from stored records carries them too. The check also compiles the installed declaration file against positive and negative type examples, including a hand-written nine-member ledger stand-in that must still be accepted by the two free functions (their ledger parameter only needs the id-keyed reader). Since 0.91.0 every delivered record the ledger writes names its source (`by`: a frame from the new run, or the accepted steer receipt) and may carry a share (`full` or `partial`, absent meaning full); a delivered record without a source, as written by earlier versions, is read as unknown rather than delivered, a record whose extension fields cannot be read falls back to that unknown form and is stored without them, and keys the reader does not know are dropped and counted. Merging takes the highest of unknown, partial and full for each message, whatever the order: the check replays 2400 random sequences in original, shuffled, duplicated, split-and-merged and persisted forms against an independent model, and feeds records written by this version to the previous release's reader, which must give the same answers. A submission can also carry the engine generation it was sent on; the ledger then lists, once per generation, the messages submitted there that are still known undelivered, without changing the recorded facts, and a malformed generation is ignored while the submission itself is still recorded. Revoked proxies passed as batches, outcomes, record lists or message lists are treated as unreadable instead of throwing. Both write ports also take the two tiers of an encoded submission, `{ full, partial }` as returned by `deliveredUuidsOf`: the partial tier is recorded with `share: 'partial'`, so a message that was only partly sent is never read back as known undelivered, and an unreadable batch records nothing without throwing. On read-back, once a record's `u` and `s` are read, a throwing extension field or key enumeration keeps the record as legacy instead of dropping it. |
 | `scripts/run-session-memory-erase-projection-test.mjs` | The session owner's memory erase (`eraseSessionMemory`, with the pure judge `sessionMemoryEraseOutcome`): one call sends exactly one request to erase what a session contributed to memory, with a body of exactly the request id plus the unevidenced-erase consent only when the caller gives it as a boolean (never added when absent, and no other key the caller passes reaches the wire); a dot-segment session id is refused locally. The connection takes the same configuration as the engine client factory, and the outbound headers are compared byte for byte with a client that factory builds for every credential form (string, per-request value function, a value function that returns nothing or throws, empty, same-origin relay, loopback without a token, blank or missing principal); a token-bearing connection in a browser host sends nothing. Malformed arguments, options and connections never throw and send nothing. A request that may have reached the engine without a readable answer (time limit, a connection dropped after the request went out or while the body was read, a cancellation after sending, an unreadable response) is reported as no answer, while a connection that never opened, a cancellation before sending or a synchronous fetch failure is reported as not sent. The judge sorts each answer into erased, partly erased, refused (nothing was erased: not sent, or a named engine code issued before any change) or not confirmed (no answer, an unreadable receipt or one for a different request or session, a refusal with two possible meanings, an unrecognised code, or a non-success status without the engine's own error code; a code found only on the prototype chain does not count), by machine code before HTTP status, with one fixed sentence per outcome and a retry posture; a caller who gave the unevidenced consent always gets a person decides on the not confirmed outcomes. Receipts are read with the existing erase receipt reader. The closed lists are frozen, derived from the wording tables and equal to what the judge produces; sentences differ, refusals always say nothing was erased, and none copies a server sentence. The check also boots installed server packages of the current and the previous generation, when available, with their real dispatcher and handler and the engine's real file memory store, and drives the call against them: real receipts and replays, a request id reused on another session, a store without erase evidence with and without consent, unknown and foreign sessions, missing credentials, missing faces, malformed bodies and the coded refusals; the previous generation's untyped internal errors are judged not confirmed. The unevidenced-erase consent counts only when it is the caller's own property (an inherited one is never sent), a session id that cannot be encoded into the path is refused locally without a request and without a rejection, and a signal-shaped object that is not a real abort signal is refused as bad input. |
 | `scripts/run-disable-skills-wire-test.mjs` | The request field that turns off every skill for one request (the deployment's skills, skills the client sends and skills listed on agent definitions). It is carried on the two user lanes only as the literal true; false or an absent value sends nothing, and any other value is refused before anything is sent. A request that also carries a non-empty skill list, or an agent definition with skills, is refused by name both when it is built and again after agent definitions are added later, because the engine rejects that combination. The value reaches a request only through a helper that reads the engine's capability report: when the engine has not reported the capability, reports it as unavailable, has not been read yet, or reports it in a form that cannot be read, nothing is sent (an engine that does not know the field rejects the whole request) and the caller receives the reason plus one sentence saying skills stay on for this run. Requests that do not use the field are byte-identical to the release before it was introduced, apart from images on the one-shot lane, which gained a seat in a later release and are checked by the request-receipt and human-submission guards instead. |
-| `scripts/run-permissions-declaration-test.mjs` | The request-side assembly of permission rules (`permissionsDeclaration`): the deny, ask and allow lists the client resolved from its settings become the `settings.permissions` fragment of a request. Every list is ordered by where each rule came from (managed policy first, then command-line flags, the settings flag, the session, user, local and project settings, and rules of unknown origin last) and filtered by the same judges the engine uses, so a rule the engine would refuse does not turn a whole request into an error — the judges are the client's installed engine functions, so the server can still refuse the whole request when its engine version differs from the client's, when a judge fails on one rule and that deny or ask rule goes through unchecked, or when a list is longer than the engine's limit. A whole-server or whole-peer rule (`mcp__<server>`, `a2a__<peer>`; one rule for every protocol prefix the installed engine knows) is rewritten to the same name followed by `__*`, after checking, protocol by protocol, that the engine judges both forms alike and that every tested server generation accepts the rewritten form. A whole tool name that the engine's rule compiler refuses (for example one containing a closing parenthesis or a backslash) is kept on the client as well, whether or not the rule was rewritten, so every rule sent has passed the same judge; this is checked on generated rule strings and differs from the previous terminal's behaviour in exactly that one registered class. Allow rules are sent only when the engine reports, through one capability position read by `observedPermissionsAllowGrant`, that it reads allow rules as rules rather than as an exclusive list; an engine that does not say so may read an allow list as the only tools permitted, so nothing is sent and approvals keep being answered on the client side. A sent allow rule records that someone answered for the tool; it does not waive the prompts the engine raises on its own, so the client keeps answering matching approvals from all of its allow rules, and allow rules with a content pattern are never sent. Allow rules that come from settings files (or whose source is unknown) are also held back whenever the host does not report that the settings read which produced them was clean (`settingsRead: 'clean'`; a partial or unreadable read, or no report at all, holds them back), while allow rules from the command line or the session are still sent, so every lane follows one rule decided in one place. The capability position is the one servers publish from 7.115.0 on, checked against the capabilities route of each server generation the gate is given. An empty list is never sent, missing rule judges raise an error instead of sending unchecked rules (a judge that is present but fails on a rule is the exception above), the diagnostic sentence is derived from the same judgement as the sending decision and never throws, and inputs with inherited or malformed fields, or rule lists whose gaps are filled from a modified array prototype, are refused rather than silently losing a rule. A companion report, `permissionsDeclarationReport`, comes out of the same single pass: its `permissions` are byte-identical to the sending decision, every rule string lands exactly once either in what is sent (with a table mapping each rewritten form back to the rule that produced it) or in a `kept` list with one reason from a closed vocabulary, and deny or ask rules sent while the rule compiler could not answer are listed as unchecked; the gate keeps one witness per reason, checks that both entry points make the same sequence of compiler calls and throw the same errors, and compares the sending decision with the previous release on two independent rule pools. From 0.92.0 the allow list's whole tool names are checked under the tightening reading, because the engine's rule compiler no longer takes an allow reading at all: the guard proves the swap changes nothing on the wire — the published 0.90.0 build running on the engine generation that receives this list and the current build running on either engine generation emit byte-identical requests over two rule pools, every name the receiving generation refuses under the allow reading is refused by the current compiler under the deny reading and no other, and the mirrored rule shape's behavior set equals the installed compiler's. It also records why the engine's own allow-list grammar is not used as the judge yet: on the names this assembly checks, that grammar and the receiving generation's compiler disagree in exactly five registered classes (names with a backslash, which the receiving generation refuses and would reject the whole request for, are never sent), and a sixth class appearing turns the guard red. From 0.92.0 it also pins how path rules relative to the task root (`src/**`, `./x`, `/x`) are handled: they are sent exactly as written, byte-identical to every published build from 0.90.0 on (0.91.0 included, whose companion report is compared in full as well whenever the capability that turns allow rules on is not reported as true), and never rewritten to an absolute `//` form; for every server generation the gate is given, it records verbatim how that server's settings intake answers such deny and ask rules with and without a task root (refused without one, accepted with one, absolute `//` rules accepted either way), so a change on the server side turns the guard red. From 0.92.0 two of those classes (protocol names whose server segment carries a backslash or a wildcard) are held back before the name is checked at all, so three classes reach that judge, and the guard checks each held witness was never asked. Protocol rule names whose server segment the engine never mints as written (a wildcard segment, or one the engine's own naming function would rewrite) match no tool on any server generation: in the allow list they are held back with the reason `unminted_protocol_segment`, while deny and ask rules are still sent exactly as the previous release sent them (a tightening rule must not quietly disappear; from 7.116.0 the server refuses the whole request and names the rule) and are listed in the report's `unreachable` table with the prefix the engine mints for that server, so a client can warn before sending; the guard compares the deny and ask lists with the previous release byte for byte on a pool full of such names, and checks that the table lists exactly the sent rules the server's judge would refuse. Without the naming function installed the allow copies are held as not judged and only wildcard segments are listed. Since 0.94.0 the allow bucket follows the engine generation that receives it: with the allow-layer capability (and the engine's allow-list reader installed) content-form allow rules are sent as written, and an entry the engine would read as voiding the whole allow list is held back and reported; older engines keep the previous projection byte for byte. |
-| `scripts/run-file-history-leg-notice-test.mjs` | What a client says after the engine reports, for one run, whether it is capturing file history. A five-way verdict combines the engine's capture capability, whether this run's submitted request actually carried the per-run opt-out, whether the user turned file checkpoints off, and the word reported for the run; only one outcome produces a sentence (the run carried the opt-out and the engine still captures), with its conclusion first so a truncated status line still reads. The guard pins every one of the 54 cells of the verdict table, sweeps alternative spellings and malformed values for each input, checks that an unreadable capability never produces a sentence while a carried opt-out is judged without consulting the capability, and that inherited input properties are ignored. A process-local ledger records, per submission, whether the outbound request body carried the opt-out (only a plain own data property counts; a body with its own serialisation hook or an accessor for that key is read as unknown, so a fact is never recorded that the sent JSON could contradict) and binds the submission to its run only on a frame proving the run was created, using the same rule as the prompt-delivery ledger, so session-busy refusals and replayed approval cards carrying another run's id never attach this submission's fact to that run; answers do not depend on call order, conflicting records read as unknown, and runs this process never bound read as unknown. An end-to-end section sends real requests through the SDK stream parser to a local stub and drives the reported word through the projection and adapter layers to the verdict. A body whose property descriptor and property read disagree (a proxy), or that carries any enumerable getter, reads as unknown, because serialization runs those getters and may drop the field; a value that is a non-null object (a wrapped string, an object with its own serialisation hook) also reads as unknown, since what it serialises to cannot be proved. |
-| `scripts/run-support-floor-projection-test.mjs` | The package's machine-readable support floor: the oldest server, engine and SDK versions it still supports, kept in `scripts/support-floor.json` with the source of each value and exported at runtime as `SUPPORT_FLOOR` (with `supportFloorOf` and the engine version reading `engineVersionTier`). The check fails when the declaration cannot be read or is malformed, when the exported values differ from it, when the SDK value differs from the SDK peer range, and (when the full fixture set is required) when no server fixture in the list matches the declared server version or that server pins and installs a different engine than declared; every fixture in the list is printed with whether it sits below, at or above the floor. The version reading answers engine at and above the floor, legacy below it and unknown for anything that is not a plain release version (pre-releases, build metadata, a leading v, whitespace, wrong segment counts, leading zeros, the word unknown, non-strings); a caller-supplied threshold can only raise the floor, a malformed one is refused before any version is read, and hostile inputs never throw. The same reading exists for the version of the engine install's core package (`coreVersionTier`): the same parser, the same three answers and the same result type, differing only in where the lower bound comes from (the declared engine floor rather than the declared server floor); the check compares the two readings on one pool of inputs, pins the golden cases, and fails when either reading carries its own version comparison instead of calling the one shared function. Since 0.93.0 the check also pins that exactly one function in the whole source tree reads a version string (a planted version-shaped regular expression or split-and-parse is detected by the same scan); the per-feature version checks that used to carry their own thresholds — the limits and detach checks, the goal stop-hook tier, the plan-review three-choice tier and the inherited-environment server tier — no longer exist. |
+| `scripts/run-permissions-declaration-test.mjs` | The request-side assembly of permission rules (`permissionsDeclaration`): the deny, ask and allow lists the client resolved from its settings become the `settings.permissions` fragment of a request. Every list is ordered by where each rule came from (managed policy first, then command-line flags, the settings flag, the session, user, local and project settings, and rules of unknown origin last) and filtered by the same judges the engine uses, so a rule the engine would refuse does not turn a whole request into an error — the judges are the client's installed engine functions, so the server can still refuse the whole request when its engine version differs from the client's, when a judge fails on one rule and that deny or ask rule goes through unchecked, or when a list is longer than the engine's limit. A whole-server or whole-peer rule (`mcp__<server>`, `a2a__<peer>`; one rule for every protocol prefix the installed engine knows) is rewritten to the same name followed by `__*`, after checking, protocol by protocol, that the engine judges both forms alike and that every tested server generation accepts the rewritten form. A whole tool name that the engine's rule compiler refuses (for example one containing a closing parenthesis or a backslash) is kept on the client as well, whether or not the rule was rewritten, so every rule sent has passed the same judge; this is checked on generated rule strings and differs from the previous terminal's behaviour in exactly that one registered class. Allow rules are sent only when the engine reports, through one capability position read by `observedPermissionsAllowGrant`, that it reads allow rules as rules rather than as an exclusive list; an engine that does not say so may read an allow list as the only tools permitted, so nothing is sent and approvals keep being answered on the client side. A sent allow rule records that someone answered for the tool; it does not waive the prompts the engine raises on its own, so the client keeps answering matching approvals from all of its allow rules, and allow rules with a content pattern are never sent. Allow rules that come from settings files (or whose source is unknown) are also held back whenever the host does not report that the settings read which produced them was clean (`settingsRead: 'clean'`; a partial or unreadable read, or no report at all, holds them back), while allow rules from the command line or the session are still sent, so every lane follows one rule decided in one place. The capability position is the one servers publish from 7.115.0 on, checked against the capabilities route of each server generation the gate is given. An empty list is never sent, missing rule judges raise an error instead of sending unchecked rules (a judge that is present but fails on a rule is the exception above), the diagnostic sentence is derived from the same judgement as the sending decision and never throws, and inputs with inherited or malformed fields, or rule lists whose gaps are filled from a modified array prototype, are refused rather than silently losing a rule. A companion report, `permissionsDeclarationReport`, comes out of the same single pass: its `permissions` are byte-identical to the sending decision, every rule string lands exactly once either in what is sent (with a table mapping each rewritten form back to the rule that produced it) or in a `kept` list with one reason from a closed vocabulary, and deny or ask rules sent while the rule compiler could not answer are listed as unchecked; the gate keeps one witness per reason, checks that both entry points make the same sequence of compiler calls and throw the same errors, and compares the sending decision with the previous release on two independent rule pools. From 0.92.0 the allow list's whole tool names are checked under the tightening reading, because the engine's rule compiler no longer takes an allow reading at all: the guard proves the swap changes nothing on the wire — the published 0.90.0 build running on the engine generation that receives this list and the current build running on either engine generation emit byte-identical requests over two rule pools, every name the receiving generation refuses under the allow reading is refused by the current compiler under the deny reading and no other, and the mirrored rule shape's behavior set equals the installed compiler's. It also records why the engine's own allow-list grammar is not used as the judge yet: on the names this assembly checks, that grammar and the receiving generation's compiler disagree in exactly five registered classes (names with a backslash, which the receiving generation refuses and would reject the whole request for, are never sent), and a sixth class appearing turns the guard red. From 0.92.0 it also pins how path rules relative to the task root (`src/**`, `./x`, `/x`) are handled: they are sent exactly as written, byte-identical to every published build from 0.90.0 on (0.91.0 included, whose companion report is compared in full as well whenever the capability that turns allow rules on is not reported as true), and never rewritten to an absolute `//` form; for every server generation the gate is given, it records verbatim how that server's settings intake answers such deny and ask rules with and without a task root (refused without one, accepted with one, absolute `//` rules accepted either way), so a change on the server side turns the guard red. From 0.92.0 two of those classes (protocol names whose server segment carries a backslash or a wildcard) are held back before the name is checked at all, so three classes reach that judge, and the guard checks each held witness was never asked. Protocol rule names whose server segment the engine never mints as written (a wildcard segment, or one the engine's own naming function would rewrite) match no tool on any server generation: in the allow list they are held back with the reason `unminted_protocol_segment`, while deny and ask rules are still sent exactly as the previous release sent them (a tightening rule must not quietly disappear; from 7.116.0 the server refuses the whole request and names the rule) and are listed in the report's `unreachable` table with the prefix the engine mints for that server, so a client can warn before sending; the guard compares the deny and ask lists with the previous release byte for byte on a pool full of such names, and checks that the table lists exactly the sent rules the server's judge would refuse. Without the naming function installed the allow copies are held as not judged and only wildcard segments are listed. Since 0.94.0 the allow bucket follows the engine generation that receives it: with the allow-layer capability (and the engine's allow-list reader installed) content-form allow rules are sent as written, and an entry the engine would read as voiding the whole allow list is held back and reported; older engines keep the previous projection byte for byte. The one-sentence notice for the print lane when settings-file allow rules were held back because the settings read was not clean (`permissionsAllowKeptNotice`) is pinned here: the wording per read-health state, the managed-policy variant, no rule text in the sentence, malformed input treated as absent. |
+| `scripts/run-file-history-leg-notice-test.mjs` | What a client says after the engine reports, for one run, whether it is capturing file history. A five-way verdict combines the engine's capture capability as it was read when the request was assembled (not as it reads when the report arrives), whether this run's submitted request actually carried the per-run opt-out, whether the user turned file checkpoints off, and the word reported for the run; only one outcome produces a sentence (the run carried the opt-out and the engine still captures), with its conclusion first so a truncated status line still reads. The guard pins every one of the 54 cells of the verdict table, sweeps alternative spellings and malformed values for each input, checks that an unreadable capability never produces a sentence while a carried opt-out is judged without consulting the capability, and that inherited input properties are ignored. A process-local ledger records, per submission, whether the outbound request body carried the opt-out (only a plain own data property counts; a body with its own serialisation hook or an accessor for that key is read as unknown, so a fact is never recorded that the sent JSON could contradict) and binds the submission to its run only on a frame proving the run was created, using the same rule as the prompt-delivery ledger, so session-busy refusals and replayed approval cards carrying another run's id never attach this submission's fact to that run; answers do not depend on call order, conflicting records read as unknown, and runs this process never bound read as unknown. An end-to-end section sends real requests through the SDK stream parser to a local stub and drives the reported word through the projection and adapter layers to the verdict. A body whose property descriptor and property read disagree (a proxy), or that carries any enumerable getter, reads as unknown, because serialization runs those getters and may drop the field; a value that is a non-null object (a wrapped string, an object with its own serialisation hook) also reads as unknown, since what it serialises to cannot be proved. |
+| `scripts/run-support-floor-projection-test.mjs` | The package's machine-readable support floor: the oldest server, engine and SDK versions it still supports, kept in `scripts/support-floor.json` with the source of each value and exported at runtime as `SUPPORT_FLOOR` (with `supportFloorOf` and the engine version reading `engineVersionTier`). Since 0.95.0 the constant is not a hand-kept copy of the declaration but its build-time product: a generator writes one comment-free, import-free file from the declaration, and the check fails when the checked-in file differs from a fresh generation, carries a comment, when the runtime constant is not that very object, or when the source holds a literal version string; it proves itself against an unfaithful generator (a dropped key, a changed key order, a stale file, an injected comment, an orphan or missing file) and against twelve malformed declarations. The check also fails when the declaration cannot be read or is malformed, when the exported values differ from it, when the SDK value differs from the SDK peer range, and (when the full fixture set is required) when no server fixture in the list matches the declared server version or that server pins and installs a different engine than declared; every fixture in the list is printed with whether it sits below, at or above the floor. The version reading answers engine at and above the floor, legacy below it and unknown for anything that is not a plain release version (pre-releases, build metadata, a leading v, whitespace, wrong segment counts, leading zeros, the word unknown, non-strings); a caller-supplied threshold can only raise the floor, a malformed one is refused before any version is read, and hostile inputs never throw. The same reading exists for the version of the engine install's core package (`coreVersionTier`): the same parser, the same three answers and the same result type, differing only in where the lower bound comes from (the declared engine floor rather than the declared server floor); the check compares the two readings on one pool of inputs, pins the golden cases, and fails when either reading carries its own version comparison instead of calling the one shared function. Since 0.93.0 the check also pins that exactly one function in the whole source tree reads a version string (a planted version-shaped regular expression or split-and-parse is detected by the same scan); the per-feature version checks that used to carry their own thresholds — the limits and detach checks, the goal stop-hook tier, the plan-review three-choice tier and the inherited-environment server tier — no longer exist. |
 | `scripts/run-server-7113-fixture-test.mjs` | Adoption checks against the published server package, run once per server generation (the one that carries the change and the one just before it). Session reads, live steer and policy settlement are driven through the package's real routes and modules; the attachment checks read the package's shipped source text instead. Attachments: the keys the server's shipped shape description lists are exactly the keys this package knows, the server's shipped refusal for the retired plan-mode reminder key names its retirement, every value this package's attachment switch can produce stays inside the accepted set, and naming the retired key is refused locally instead of being sent. Session reads: a non-owner reading a session's detail, head or event stream gets a not-found whose code names the session (the previous generation named a run), and this package's session read answers the same way on both. Live steer: without an idempotency key the engine's input id equals the message id on the receipt, a queued frame and an applied frame on one input id each project as their own event with every field as written, and the delivery ledger records delivery once. Policy settlement: an unattended answer by class is recorded under the policy party as an allow settled by policy, read back verbatim, no longer read as a person's approval, and a tool-end frame carrying it still projects. |
 | `scripts/run-server-7114-fixture-test.mjs` | Adoption checks against the published server package's real task admission, run on every configured server package (the newest one and each older one in the list), each held to an exact expectation for its generation. From 7.114.0, a tool-event hook matcher that selects a tool's alias but not its current name is refused with a field-invalid error naming both names, in every spelling the server reads (literal, pattern, padded, comma list, pipe list, empty segment), and a matcher that also selects the current name is accepted; 7.113 refuses only the literal and the pipe-list spellings; 7.111 and older accept all nine. A server generation outside the table fails the check instead of being guessed, and with fixtures required each generation band must be covered. This package passes the matcher through byte for byte, the request it assembles gets the same answer from the server as the hand-written one, and its local skip check keys on the matcher as written. From 7.115.0 the refusal additionally carries a machine-readable `field` and a `detail` naming the event, the alias and the current name, with the sentence and the error code unchanged; older generations carry neither. |
 | `scripts/run-steer-refusal-advice-test.mjs` | Refusals of a mid-run steer have one speaker. The root entry's `steerRefusalAdviceOf(error)` returns the steering refusal's safety code and disposition sentence, and `ControlRouter.steer` builds its `ControlSafetyError` from that very call, so a client that steers without the router (and therefore holds the raw SDK error) shows the same words the router would instead of keeping its own sentence table. The guard runs a pool of 180 thrown values (every steering code under ten status shapes, every SDK class name, non-steering errors, throwing getters, proxies, inherited fields and non-objects) through both paths and requires the same code, sentence, message, status and cause on every row; it checks every message against the bytes the previous release produced on the same pool, recorded from that build rather than hand-written, with the four rows whose field reads throw registered as the one intended change: the original error now propagates instead of the getter's. It also pins the error's two new read-only fields (`advice`, `status`), absent on the old two- and three-argument constructor calls and on the cancel 404 and local refusals that carry no disposition; the call-site roster of the three steering legs; that each sentence exists exactly once in the source; and that the open-set sentence splits on the same 4xx line the delivery verdict uses. A last section compares each sentence with the table a client kept, so a change on either side forces a fresh reading. From 0.92.0 the sentences are rewritten so a client can show them as they are, each is built from two halves the result also carries, and the cancel 404 gains one: the guard keeps comparing the code and the message prefix with the previous recorded bytes and requires the message to be that prefix followed by the new sentence |
@@ -475,26 +475,26 @@ guard still cross-checks the table by name).
 | `scripts/run-hook-matcher-verdict-test.mjs` | The hook-matcher admission verdict `hookMatcherVerdict(event, matcher)` against the published server package's real task admission: for 35 matcher shapes on each tool event, a `refused` verdict (the matcher selects a tool alias but not that alias's canonical name) matches a 400 from the server, and `accepted` / `matchesAll` match an admitted request; from server 7.115.0 the refusal's machine-readable `detail` (event, alias, canonical) must equal the verdict's pair. Shapes the verdict does not judge are listed with their reason and are never treated as accepted. The guard also pins that the verdict reads the generated alias table (adding a pair to a copy of the table changes the copy's verdict), that only the three tool events are judged, that the verdict is case-sensitive and never throws on non-string input, and that pathological user regexes return within a bounded time (each call runs in a worker with a deadline; the same regex run directly against the longest tool name does not finish). After the fact, `hookMatcherRefusalFromError` reads the server's own refusal: from 7.115.0 the refusal's `field` and `detail` give the event, the alias and the current name, which must equal this function's verdict on the same matcher; older refusals carry no such fields and the reader answers nothing rather than parsing the sentence. |
 | `scripts/run-human-turn-encoding-test.mjs` | The check for the data envelope that carries a user's @-attached file, notebook, image, PDF or directory to an engine that has no attachment seat, and for the single function that turns one human submission into what the request carries. The envelope bytes are pinned against the 45 reference samples agreed with the terminal (the sample file is a byte-for-byte copy), and four extra samples with an instruction inside the file contents are checked for shape: the instruction stays between the two fence lines with a line-number prefix, the fence is longer than any tilde run in the envelope, and the header carries no reminder tag, no claim that a tool was called and no imperative. The three boundary rules (fence length over the whole envelope including look-alike tildes, visible escapes for in-line break characters, quoting and escaping of names) each have their own cells; a malformed descriptor throws and accessor properties are never run. Every envelope is wrapped in a closed-set `<attachment>` tag carrying a `path` attribute (a reference sentence gets a self-closing tag line instead); the text inside the tag is pinned byte for byte against the envelopes minted before the wrap, 26 frozen fixtures covering every descriptor kind; the attribute is the shown path with the four markup characters escaped, checked on hand-written cases and a seeded pool; descriptors without a path get a tag without attributes; and each envelope reads as exactly one opening line and one closing line. For the assembly: the projection rule (non-empty items joined by a newline, attachments before the item text), the image seat plan (pasted images first, at most 20 images, each image up to the engine's inline-image limit and 6 MiB in total, only when the model can see images), the 8 MiB request-body accounting (exactly at the limit is accepted and one byte more degrades or refuses, compared with the real serialized body on six body shapes), the order in which attachments are reduced to a reference sentence, the three-step handling on the follow-up carrier (contents, then reference sentences, then keep queued) and its agreement with the engine's own follow-up text validator on a seeded pool, the delivery shares and the reduced list as two views of one record, the refusal reasons, the image-vision verdict and the attachment-seat capability reader. The two engine limits the package reads from generated constants (the per-image size and the follow-up text length) are checked against the engine's own exports, and the per-image size also against the engine's inline-image admission path; the limits it still mirrors are checked against a server package fixture's real admission path when one is configured, and every server fixture generation must admit an image of the per-image limit. Pasted images take seats first but are held to the same count, per-image and total limits; an image over a limit is left out and recorded as not sent. Very long rider lists are handed to the ledger without overflowing the call stack, and an encoded submission handed to the ledger as-is records partly sent entries as partly delivered. |
 | `scripts/run-server-7115-fixture-test.mjs` | Checks against the published 7.115.0 server package and every older package in the list, each held to an exact per-generation table. A server tool call nobody has spoken for is classified as the new `unstated` ask class by the server's own engine and frame reader, and that pair reaches this package's card and seat unchanged, with the standing-answer verdict matching the engine's own arbitration row. The capabilities route really runs: from 7.115.0 it reports the allow-rule capability and this package reads it as present, older servers do not report it and the allow bucket is not sent; requests this package assembles from a mixed rule pool are accepted by the same generation's admission and settings compiler, while content-form allow rules sent raw are refused there. The server's refusal for tasks parked before an upgrade is produced by its own function and, through the SDK error mapping, is read by this package's dedicated reader and turned into a three-part sentence that says the work is still waiting, with the way out chosen by the parked kind the caller supplies; refusals without the flag are not claimed. The two continue-only-an-existing-conversation refusals are read the same way on every generation, and the prose probe reason minted for live frames equals the durable row's copy and lands on the card from both legs. |
-| `scripts/run-replayed-approval-disposition-test.mjs` | Approval requests read back from a run's event log are only raised as questions when the pending-approval list says they are still open: `replayedApprovalDisposition(frame, basis)` answers `live` / `decided` / `unknown` (a frame without an event-log position is live; one listed by this connection's opening snapshot or by a pending list read after the frame arrived is live; one missing from a complete list is decided, with the crash-convergence cause when the same response carries one; an incomplete, missing or unreadable list, or no read, is unknown and never treated as open), and the in-stream approval leg applies the same verdict before raising a card — a decided frame raises no card and sends no decision, an unknown one raises no card, sends nothing, hands the frame to `onReplayedApproval` and shows one notice, the list is re-read for every such frame (never cached across frames), the read is bounded, and the opening-snapshot evidence is reset on every new connection. |
+| `scripts/run-replayed-approval-disposition-test.mjs` | Approval requests read back from a run's event log are only raised as questions when the pending-approval list says they are still open: `replayedApprovalDisposition(frame, basis)` answers `live` / `decided` / `unknown` (a frame without an event-log position is live; one listed by this connection's opening snapshot or by a pending list read after the frame arrived is live; one missing from a complete list is decided, with the crash-convergence cause when the same response carries one; an incomplete, missing or unreadable list, or no read, is unknown and never treated as open), and the in-stream approval leg applies the same verdict before raising a card — a decided frame raises no card and sends no decision, an unknown one raises no card, sends nothing, hands the frame to `onReplayedApproval` and shows one notice, the list is re-read for every such frame (never cached across frames), the read is bounded, and the opening-snapshot evidence is reset on every new connection. The test for 'does this frame carry an event-log position' is public (`hasLedgerPositionKey`: either an `id` or a `seq` key that is present and not null counts, whatever its value; it is the exact opposite of the verdict's live-leg shortcut), and the one sentence a host shows for a decided or unknown frame is minted in one place (`replayedApprovalSentence`): a decided frame is described only as no longer listed by the engine as waiting (a tool name, when given, is sanitised and shown inside quotation marks), a crash-convergence cause says the engine denied it automatically (it stopped unexpectedly, or was shut down in an orderly way), nothing is ever described as approved or as having run, an unknown frame says it could not be confirmed, and every sentence says no decision was sent. A question the bridge has already seen settled (answered, dismissed with an interrupt, or closed by an engine closing frame) is never raised again by a repeated frame — with or without an event-log position, whether or not the opening snapshot lists it, and even when the pending list read afterwards still lists it: the verdict is `decided` with a `settled` word (never together with a crash-convergence cause), no card, no decision, no list read, no notice, and one hand-off to `onReplayedApproval`. Only certain outcomes are recorded (a card the host withdrew or a decision that failed to send is not, so a repeated frame still raises a card then), the record is first-outcome-wins, survives new connections, is bounded, and the sentence for it is minted in the same one place without saying what the outcome was. |
 | `scripts/run-server-7116-fixture-test.mjs` | Checks against the published 7.116.0 server package and every older package in the list, each held to an exact per-generation table. Answering a stream approval that the server has already closed without a human decision (a 410 `gone.ask`, which from 7.116.0 also covers an approval left pending by a process that crashed or shut down) is produced by the server's own convergence step and decision route, and through both SDK error mappings is read by `askGoneFromError` and turned by `askGoneContent` into a three-part sentence whose way out depends on the cause; the decision-failure note and the already-resolved check claim it by code; the two process-ended causes are the same closed list as crash-converged rows, the sentence carries no machine words, and the one denied-group word the reader keeps equals the server's table. Run death codes from the server's run-detail route reach the abort-cause reader and the synthesized terminal unchanged, with one way-out sentence per class. An empty operator list now answers 501 `capability.operator_principals_required`, read as `operators_unconfigured` (nothing was done) by the memory and rules readers and the rule write, while 403 stays `operator_only`; a plan-review decision hitting an unavailable-face 501 is `not_applied`; the four new capability bits are read by their readers. Protocol rule names whose server segment the engine never mints are held back from the allow list, while deny and ask rules are sent unchanged and listed in the report's `unreachable` table; the server's own judge refuses exactly the listed rules (naming the same minted prefix), the request without them is accepted, and the whole request is refused by 7.116.0 and accepted by older servers. The remember-refused card flag reaches the card from both paths. |
 | `scripts/run-automatic-answer-verdict-test.mjs` | The single verdict `automaticAnswerVerdict(card, lane, facts?)` that decides whether an automatic lane (the standing permission mode, or a standing allow rule) may answer an approval card on the user's behalf. The only `may_answer` exit is a well-formed class set whose classes are all known to this build, whose folded row does not block on that lane, with no must-ask evidence on the card and, for the mode lane, a host reading that its own ask rules did not match; every other card is `must_ask` (with the road that decided it) or `unknown` (with why). The guard derives every single-class card and every ordered pair of classes on both lanes from the engine's own arbitration table, walks every row of the decision table on both sides (including positive evidence when the class set is absent or malformed, class sets that change while being read, and class names this build does not know), compares the package's fold against the engine's fold function over all singles, pairs, triples and a random sample, checks the mode lane against the engine's bypass-mode arbitration (including the release disclosure), confirms on a real engine run that a matched ask rule always puts `ask_rule` in the class set, checks that the two older predicates never answer more loosely than this verdict nor than the previous release, and, when a terminal reference tree is available, that the terminal's own decision chain auto-approves nothing new when it is built against this package. Since 0.94.0 it also covers the class-aware form of the mandated-question sentence: the clause naming an automatic check appears only when the card's question classes let the classifier or the sandbox answer, checked class by class against the engine's own table, and any unreadable card falls back to the generic sentence. Since 0.94.0 the mode lane reads the bypass column on the tool face of the card (the face comes only from the generated shell-face name list, matched exactly, and releasable classes are read per face from the generated table): a write whose protection could not be resolved is answerable on the shell face and still asked on file tools, while lookalike tool names (case, whitespace, zero-width or homoglyph characters, protocol prefixes, rule syntax, inherited or accessor names, non-strings, absence) always read as the other face; when the installed engine exposes faces, the fold and the bypass arbitration are compared per face against the engine and every catalog name is checked against its face. A third lane, `permission_request_hook`, reads the hook column of the same table with the same must-ask evidence first, ignores allow-rule shadow evidence and host facts, and `permissionRequestHookScope` turns its verdict into allow-or-deny, deny-only or skip. The terminal-chain comparison accepts exactly the shell-face release as the one intended widening. |
-| `scripts/run-allow-rule-shadow-projection-test.mjs` | The reader for the fact that a standing allow layer matched a call but could not clear its ask (`readApprovalAllowRuleShadow`, `approvalAllowRuleShadowed`, and the structured `allowRuleShadow` disclosure stamped on the approval card). Newer servers send a list of members with their matched entries; older servers and older parked rows send a single legacy text, and both generations are read by the same reader (the newer key wins when both are present). The legacy card field is derived from the same reading for one more release. The guard covers every member, several members in order, empty entry lists, the count of omitted entries, member names this build does not know, four kinds of malformed values (all of which still count as evidence), absence, the legacy text on both card paths and the parked-ask path through the real card port, byte-identical answers against the previous release on legacy-shaped frames and rows, real frames and durable rows minted by the next server generation when its fixture is available, and the effect on both automatic lanes. |
+| `scripts/run-allow-rule-shadow-projection-test.mjs` | The reader for the fact that a standing allow layer matched a call but could not clear its ask (`readApprovalAllowRuleShadow`, `approvalAllowRuleShadowed`, and the structured `allowRuleShadow` disclosure stamped on the approval card). Newer servers send a list of members with their matched entries; older servers and older parked rows send a single legacy text, and both generations are read by the same reader (the newer key wins when both are present). The legacy card field is derived from the same reading for one more release. The guard covers every member, several members in order, empty entry lists, the count of omitted entries, member names this build does not know, four kinds of malformed values (all of which still count as evidence), absence, the legacy text on both card paths and the parked-ask path through the real card port, byte-identical answers against the previous release on legacy-shaped frames and rows, real frames and durable rows minted by the next server generation when its fixture is available, and the effect on both automatic lanes. The one sentence a card shows for that fact is minted in one place (`allowRuleShadowSentence`): both wire generations give the same sentence, an absent disclosure renders nothing (it never says there is no rule), an unreadable one says so, and entries and member words arrive display-sanitised. |
 | `scripts/run-steer-tier-test.mjs` | Steer priority goes through one gate. The capability reader for `fleet.steerPriority` keeps three states, takes the word as an open set and treats `fleet: false` as not reported; its judge answers `engine` only for the exact word. The steer body has one constructor, `steerBodyOf`, used by the router and by clients that build their own request: `next` and no priority never enter the body (byte-identical to the previous releases on six call shapes), `now` and `later` enter it only when the reading is `engine`, and otherwise the constructor throws a refusal with a stable code and no request is sent. Against the 7.116.0 server package the guard runs the capabilities route, the steer route with the engine's own steer verb and an in-memory checkpoint store: the words the client admits are accepted, the words it refuses are rejected with the same field error, same-key replays stay idempotent, a real tier change is refused, and a `now` steer's cut notice carries the receipt's message id. It also pins the one sentence for a same-key replay (it does not tell the user to resend under a new key unless the text changed), the rewritten refusal sentences, the single receipt reader, the cut-notice verdict and sentence through both projection layers, the first-beat order (bounded capability wait, then reading, then send) and the engine-switch chain where an invalidated reading sends nothing |
 | `scripts/run-abort-cause-projection-test.mjs` | The cause a server records when it ends a run itself — the stream client went away, the connection dropped while a question was pending and the run could not be parked in time, or a cancel request arrived. One reader handles all four places the word appears (the run record's result, the streamed `done` result, the synchronous response body and the event log's `failed` row) and reads it only on a cancelled terminal: a word on any other outcome is contradictory evidence and is not read. A missing or malformed word reads as "not recorded" rather than as any of the three, and a word this version does not know stays distinct — reported as unrecognised with the original word, never folded into a known one. The same reading drives the cancelled branch of the run-cause classifier, the `_sema_abort_cause` key on the result frame, and the single sentence appended on a new line to both the synthetic terminal row and the result frame's first error; without a cause both stay byte-identical to a cancelled terminal that carries none. The word list and the reading are checked against every server package in the fixture list — generations that mint the cause and older ones that do not — and against the client SDK's own type for it |
 | `scripts/run-transport-failure-scene-test.mjs` | The reader that turns a failed transport call (a thrown value of any shape) into a machine-readable scene: up to four layers of the cause chain, each carrying only name, code, errno, syscall and the socket byte counters — never message text, addresses, headers or request bodies — together with whether the request is provably not sent and which kind of no-answer it is. The not-sent judgement is checked against a frozen copy of the previous release's over generated corpora (code table, syscalls, evidence depth, top-level shapes, merged multi-address errors, throwing getters and proxies); the layer records are checked against a frozen copy of the terminal's own five-layer reader on the same corpus, with three registered divergence classes and none outside them; the source is pinned to a single cause-chain walker; and real undici errors from a loopback server (connection refused, peer closed after the request, peer reset) are read end to end. |
 | `scripts/run-tick-model-and-config-drift-test.mjs` | Two additions around subagent progress ticks and the engine health probe. First, the model id a delegated sub-run was prepared with now rides on the inline progress tick (`InlineTaskTick.model`, read back from the inline stats store): it is carried only when the wire reports a non-empty string, a tick without it is byte-identical to the previous release, and absence is never read as "no model". Second, the engine's config-apply counters on the health probe are read by `readEngineConfigApplyState` (the two ordinals decide; one ordinal alone counts as absent; versions and the stale duration are informational) and the heartbeat's configHash drift is judged by `configDriftVerdict` into four words — skip, in-sync, ask, advance-baseline — so a default-model hot swap that the engine has already applied no longer raises a restart question, while a restart signal from the engine, a replaced instance, or missing counters always still asks. The guard drives the real adapter and the real health route of every server package in the fixture list against a real config-apply ledger, compares the verdict input by input against the shell's own judgement, and compares adapter output byte for byte against the previous release. |
 | `scripts/run-multicause-wording-test.mjs` | Wording guard for sentences that put several possible causes side by side. When a reader folds several causes into one reading, the sentence may only be written one of two ways: split into one sentence per cause (when this end can tell which applies), or ending with a plain admission that this end cannot tell which (when it cannot). Each site is checked twice: every cause is reached through the real entry point and renders its own sentence, or the causes are shown to collapse into the same reading and the sentence admits it. Covered: the two posture sentences that used to name an engine version, the dropped-frame sentences for an unknown frame type and a malformed one, the seventeen capability readers' not-observed sentence (none received, last one unreadable, or last one discarded when the engine connection was reset), the permission-rule cap code with its engine-minted triggers, the four session-rule not-found sentences, the three allow-rule read-health sentences, and the two sentences whose engine bit encodes two causes. A source-wide census of string literals keeps a newly added reader from copying the old shape without the admission. A second pass covers twenty-two more sites, plus two of the same shape: the per-task agents warning, the cloud model-entry sentences, the decision-receipt codes, the run-death and resume-refusal sentences, the gate-vocabulary sentences, the approval-row routing sentences, the store-posture degraded sentence, the three drop paths of the "don't ask again" check (each now names only its own cause), and the unreadable-body sentences of the persisted-rule, MCP probe and memory-provenance reads (each now names the part that failed). It also checks that the two not-reported sentences no longer read a missing key as an engine version, and that the dropped-frame default sentence keeps its opening verbatim while admitting it cannot tell which. |
 | `scripts/run-workflow-decide-refusal-test.mjs` | Reader for the two refusals the decide route returns when a workflow sub-agent's pending approval cannot be handed to a single run record: `decide.workflow_park_forked` (the same run was resumed more than once and each copy waits on this same approval; the body's `runIds` come back exactly as sent — not reordered, not deduplicated, as a frozen copy — and a malformed list drops only that one field) and `decide.workflow_parks_unreadable` (the run record that owns the approval is unreadable or predates the owner field; the body's `runId` and the optional `readableHolderRunId` are read, the latter being a hint rather than a way out). Both join the existing workflow-lane refusal reader and are read by code, never by sentence; each has one user-facing sentence that carries no machine words, says nothing was delivered and the approval is still pending, and does not promise that resuming makes the decision deliverable. Both codes also count as "the approval is still waiting" by code, so a reworded server sentence containing a word like "already" is never mistaken for an already-decided approval. The guard drives the reader with real SDK error mappings from every witnessed SDK generation, with the bodies each server generation mints (older packages are pinned to "no fork code; the unreadable code carries only `runId`"; a package that mints the new bodies is run when present and reported as skipped otherwise), and compares plain inputs that carry neither new code byte for byte against the previous release. The reader never throws on a hostile error object: a field whose getter throws, or a list that is revoked part-way through validation, is dropped on its own while the recognised refusal is still returned. |
-| `scripts/run-permission-modes-capability-test.mjs` | The permission-modes capability reader and the bit-gated `dontAsk` mode. The server's capabilities body lists the permission modes its request face accepts; the reader keeps three states (a non-empty list of non-empty words is present, a missing key is an older worker that reports nothing, any malformed shape is dropped as unreadable and reads as not observed), copies the list instead of sharing the caller's array, freezes what it stores, ignores keys that exist only on the prototype, reads the top-level key once and never throws, and its stored reading is discarded automatically when the engine's capability cache changes generation. One judge, `permissionModeAdmitted`, answers true only when the reading is present, the whole list passes the same strict rule as the reader (non-empty, every item an own non-empty string) and it contains the exact word; the headless resolver reads a hand-built reading once per resolution From 0.94.0 the shell-mode normaliser is an alias table with one row (`manual` maps to `default`) and passes `dontAsk` through by name whatever reading is given; whether `dontAsk` is sent is decided in exactly one place, the field projector, which stamps it only when that judge says yes and otherwise stamps `default` for a caller that stated a mode (or nothing for one that did not). Used together, the two produce byte-identical request fields to the previous release for every reading, and every other word passes through unchanged. The field projector never stamps `dontAsk` on its own, and the headless flag lane and environment lane both follow the same rule (the environment lane says so in one warning line when it downgrades). The request-face intent table and the deployment knob's word list are derived from the one tuple, now six words with `dontAsk` last, and equal each other |
-| `scripts/run-server-7119-fixture-test.mjs` | Checks against the published 7.119.0 server package and every older package in the list, each held to an exact per-generation table with the raw readings stored. The retired decision-body key is produced by the server's own routes (the live respond route, the durable ask route and the decide route, including the unknown-keys shape and the shape-sentence shape); from 7.119.0 every refusal body carries the retired key's name as `field`, and through both SDK error mappings the decision-refusal reader recognises it by code and field, never by sentence, while the 7.118.x bodies without a field are not claimed. The capabilities route mints six permission modes in the engine's order, the real admission accepts `dontAsk` only where that list says so (unknown words are refused with the same field error in every generation), and the loop capability reading, bit-gated normaliser, field projector and admission is closed on both sides. The store posture's degraded-reason list no longer contains `db_unreachable` from 7.119.0 (the reader still understands it for older servers), and the health body's session backend label stays a string. The new 503 for an assignment naming a scenario the deployment does not have is produced by the real gate and serialised by the real admission path and is read by code with its detail. A source scan pins that this package has no reader of the council result's output keys. The decision-attribution key is chosen by the capability bit: each generation's three decision routes (live respond, durable ask, decide) accept the key this package would send and apply it as a policy answer, the other key is refused with the retired-key error from 7.118.0 on, and on older generations the new key is refused by two routes and silently dropped by the live one (recorded as a person's answer), which is why a missing bit always means the old key; both of this package's decision legs are driven against the real routes of every generation, including the direct-door refusal (its code changed name with the key) and the single resend without the key, and the two kinds of out-of-set refusal are told apart by code. Every older package is held to its own table. The device lane in the capabilities route is minted in exactly two shapes in every generation from 7.106.0 on (the whole position `false` when no hub is mounted, or an object with `enabled: true`), and the package's one lane reader is held to each: with no hub it reads the lane as absent, with only a hub it reads a present lane whose management bit is off and whose three lane facts equal what the route minted, and with the device store, enrollment and run store plus a configured per-device in-flight cap it reads the bit as on with that cap (not the default); the whole body read back through the tee gives the same reading, and an object forged with `enabled: false` is read as unreadable |
-| `scripts/run-approval-decided-by-capability-test.mjs` | The approval decision-attribution capability reader and the way both decision legs choose their key from it. The server's capabilities body carries a constant bit saying its three decision routes accept the `decidedBy` attribution key; the reader keeps three states (the literal `true` is present, a missing key is an older worker, anything else is dropped as unreadable and reads as not observed — a constant version bit has no meaningful `false`), ignores keys that exist only on the prototype, reads the top-level key once and never throws, and its stored reading is discarded automatically when the engine's capability cache changes generation. One judge, `approvalDecidedByAdmitted`, answers true only for a reading whose own data property `kind` is `present` and whose own data property `value` is `true` (accessors are never run, even if they return the right values). A single exported function, `decisionAttributionKeyOf`, decides the attribution key for both the durable decision leg and the live respond leg (and for any sender outside the package that needs the same answer), and a syntax-tree scan pins that it is declared once, called exactly twice inside the package and that no other place there builds a policy attribution literal (with planted violations proving the scan can fail): a policy refusal goes out as `decidedBy: 'policy'` only when the reading is present, as the old `settledBy: 'policy'` for every other reading (not reported, not observed, damaged, missing, unreadable, or none supplied), and with no attribution key at all for a human refusal, an absent or out-of-set word, an allow card, a failed card port or an abandoned card; the reading is judged once per decision before the first await. The direct-door refusal (its code changed name with the key, both codes read as the same arm) is answered with exactly one resend of the same body without the attribution key, whichever of the two names it carried; a second refusal is not resent, a body that carried no key, a caller that has aborted, the retired-key refusal and every other 400 are not resent, and the classification of a later refusal follows the key actually carried by the last send. The reading supplied through the host lane reaches both legs (the live frame route and the durable park path); only an own data property of the lane counts (a prototype-polluted key, an inherited key or an accessor is treated as not supplied, and the accessor is never run), and a lane that throws when read counts as not supplied, and the bridge's whitelist copy matches each key against `'policy'`, reads only own data properties (a polluted prototype or an accessor never adds a key, including on the resend body) and writes only one of them |
+| `scripts/run-permission-modes-capability-test.mjs` | The permission-modes capability reader and the bit-gated `dontAsk` mode. The server's capabilities body lists the permission modes its request face accepts; the reader keeps three states (a non-empty list of non-empty words is present, a missing key is an older worker that reports nothing, any malformed shape is dropped as unreadable and reads as not observed), copies the list instead of sharing the caller's array, freezes what it stores, ignores keys that exist only on the prototype, reads only own data properties (an accessor is never run and reads as unreadable) and never throws, and its stored reading is discarded automatically when the engine's capability cache changes generation. One judge, `permissionModeAdmitted`, answers true only when the reading is present, the whole list passes the same strict rule as the reader (non-empty, every item an own non-empty string) and it contains the exact word; the headless resolver reads a hand-built reading once per resolution From 0.94.0 the shell-mode normaliser is an alias table with one row (`manual` maps to `default`) and passes `dontAsk` through by name whatever reading is given; whether `dontAsk` is sent is decided in exactly one place, the field projector, which stamps it only when that judge says yes and otherwise stamps `default` for a caller that stated a mode (or nothing for one that did not). Used together, the two produce byte-identical request fields to the previous release for every reading, and every other word passes through unchanged. The field projector never stamps `dontAsk` on its own, and the headless flag lane and environment lane both follow the same rule (the environment lane says so in one warning line when it downgrades). The request-face intent table and the deployment knob's word list are derived from the one tuple, now six words with `dontAsk` last, and equal each other |
+| `scripts/run-server-7119-fixture-test.mjs` | Checks against the published 7.119.0 server package and every older package in the list, each held to an exact per-generation table with the raw readings stored. The retired decision-body key is produced by the server's own routes (the live respond route, the durable ask route and the decide route, including the unknown-keys shape and the shape-sentence shape); from 7.119.0 every refusal body carries the retired key's name as `field`, and through both SDK error mappings the decision-refusal reader recognises it by code and field, never by sentence, while the 7.118.x bodies without a field are not claimed. The capabilities route mints six permission modes in the engine's order, the real admission accepts `dontAsk` only where that list says so (unknown words are refused with the same field error in every generation), and the loop capability reading, bit-gated normaliser, field projector and admission is closed on both sides. The store posture's degraded-reason list no longer contains `db_unreachable` from 7.119.0 (the reader still understands it for older servers), and the health body's session backend label stays a string. The new 503 for an assignment naming a scenario the deployment does not have is produced by the real gate and serialised by the real admission path and is read by code with its detail. A source scan pins that this package has no reader of the council result's output keys. The decision-attribution key is chosen by the capability bit: each generation's three decision routes (live respond, durable ask, decide) accept the key this package would send and apply it as a policy answer, the other key is refused with the retired-key error from 7.118.0 on, and on older generations the new key is refused by two routes and silently dropped by the live one (recorded as a person's answer), which is why a missing bit always means the old key; both of this package's decision legs are driven against the real routes of every generation, including the direct-door refusal (its code changed name with the key) and the single resend without the key, and the two kinds of out-of-set refusal are told apart by code. Every older package is held to its own table. The device lane in the capabilities route is minted in exactly two shapes in every generation from 7.106.0 on (the whole position `false` when no hub is mounted, or an object with `enabled: true`), and the package's one lane reader is held to each: with no hub it reads the lane as absent, with only a hub it reads a present lane whose management bit is off and whose three lane facts equal what the route minted, and with the device store, enrollment and run store plus a configured per-device in-flight cap it reads the bit as on with that cap (not the default); the whole body landed in the capabilities cache gives the same reading, and an object forged with `enabled: false` is read as unreadable |
+| `scripts/run-approval-decided-by-capability-test.mjs` | The approval decision-attribution capability reader and the way both decision legs choose their key from it. The server's capabilities body carries a constant bit saying its three decision routes accept the `decidedBy` attribution key; the reader keeps three states (the literal `true` is present, a missing key is an older worker, anything else is dropped as unreadable and reads as not observed — a constant version bit has no meaningful `false`), ignores keys that exist only on the prototype, reads only own data properties (an accessor is never run and reads as unreadable) and never throws, and its stored reading is discarded automatically when the engine's capability cache changes generation. One judge, `approvalDecidedByAdmitted`, answers true only for a reading whose own data property `kind` is `present` and whose own data property `value` is `true` (accessors are never run, even if they return the right values). A single exported function, `decisionAttributionKeyOf`, decides the attribution key for both the durable decision leg and the live respond leg (and for any sender outside the package that needs the same answer), and a syntax-tree scan pins that it is declared once, called exactly twice inside the package and that no other place there builds a policy attribution literal (with planted violations proving the scan can fail): a policy refusal goes out as `decidedBy: 'policy'` only when the reading is present, as the old `settledBy: 'policy'` for every other reading (not reported, not observed, damaged, missing, unreadable, or none supplied), and with no attribution key at all for a human refusal, an absent or out-of-set word, an allow card, a failed card port or an abandoned card; the reading is judged once per decision before the first await. The direct-door refusal (its code changed name with the key, both codes read as the same arm) is answered with exactly one resend of the same body without the attribution key, whichever of the two names it carried; a second refusal is not resent, a body that carried no key, a caller that has aborted, the retired-key refusal and every other 400 are not resent, and the classification of a later refusal follows the key actually carried by the last send. The reading supplied through the host lane reaches both legs (the live frame route and the durable park path); only an own data property of the lane counts (a prototype-polluted key, an inherited key or an accessor is treated as not supplied, and the accessor is never run), and a lane that throws when read counts as not supplied, and the bridge's whitelist copy matches each key against `'policy'`, reads only own data properties (a polluted prototype or an accessor never adds a key, including on the resend body) and writes only one of them |
 | `scripts/run-durable-decision-single-holder-test.mjs` | The single holder of a durable approval decision: for one approval identity (the call id of the fetched pending row, or the task id when the row has none), one decision (card, answer, decide including the package's own resends) is in flight per process, and any other caller that arrives while it is in flight, or that started reading the queue before it settled, gets a copy of that same outcome marked `joinedInFlight: true` without presenting a card, sending a decision or resending anything. Two concurrent callers produce exactly one card and one decide in both arrival orders, including the real park resolver racing a direct re-presentation of the same row (the stream finishes and both sides settle); a failed decide or a throwing card port gives both callers the same failure and clears the registration so the next call is a new decision; an abort signal from either caller interrupts that one decision: the holder's card is withdrawn and the holder alone sends exactly one interruption deny that both callers report, with identical readings whichever path holds, while an abort that arrives after the decision settled sends nothing and returns the settled outcome; different identities never join; the joining caller reports the holder's card through `onPresented` only when that card was actually handed to the card port (a throwing callback does not change its outcome) and never reports a decide operation; the copy never alters the holder's object and the flag is never written as `false`; and a syntax-tree scan pins that the joining and registration code paths contain no card, decide or notice call (with a planted violation proving the scan can fail) Since 0.94.0 the identity also carries the row instance (its creation time): when the engine parks the same call a second time for a different check, that second row is a new decision in both arrival orders (while the first card is still on screen, or when a read that started before the first decision settled returns the second row), and two callers fetching the same row instance still meet. |
 | `scripts/run-approval-respond-parking-test.mjs` | The live-card respond answer when the engine replies `202` with the reason `parking_in_progress`: the card's row is moving to parked, so the answer did not take effect. The outcome is `decision: 'unresolved'` plus a `parking` position carrying the approval id and, when the engine supplies them, the suggested retry interval and the decide path of the parked card; the resolution reader reports it on the existing unsettled arm with the new cause `parking_in_progress` (the union keeps its three arms; a retracted or edit-refused card still wins over it, and it wins over the decision word). The parking fact is carried only by an own data property naming the reason and a non-empty approval id that matches the card; a retry interval or decide path that is unusable drops only itself and the parking still stands. A body that names the parking reason but whose approval id is missing, unusable or belongs to another card is reported as not settled, never as decided and never as parked, with no notice and no resend. The answer is never re-sent, never has its attribution key dropped or renamed afterwards, and the package does not wait for the interval; no accompanying notice (for example the one saying the approval itself went through) and no attribution position is produced; exactly one user-facing sentence is shown (with the engine's suggested wait appended in seconds when given) and it contains no engine text. A 200 acknowledgement keeps its reading unchanged, bodies that are neither acknowledgements nor parking bodies are not read as parking, the durable decision path and its single holder have no parking cause, and a syntax-tree scan pins one reader declaration, one call site and the places where the reason word is minted, while the 202 body shape is pinned against every server generation in the fixture list |
 | `scripts/run-decision-refusal-standing-test.mjs` | Shared decision-refusal reading for every client: when the engine refuses an approval decision, `classifyDecisionRefusal` classifies the failure only by status and machine code into settled elsewhere (404, 410, or a 409 `conflict.ask_decided`), ambiguous (5xx except 501) or refused while the question is still pending (every other 4xx and 501), and returns nothing when no server verdict can be read, and marks transient refusals (408, 425, 429, 502, 503, 504, or a body that says the request is retriable); `liveAskPendingReading` reads whether the question is still in the engine's live pending list as pending, not pending or unknown (an unreported, malformed or partly dropped list is unknown); `decisionRefusalClosesOut` closes out only a refusal that is neither settled elsewhere nor transient and whose question the read confirms is still pending; `undeliverableDecisionContent` is the one sentence for each way the close-out ended. The guard covers the golden status/code table and near-miss codes, values that are not a verdict, guarded reads (throwing getters, prototype accessors, revoked proxies), the outcome note shape, codes unsafe for display, the full close-out truth table and the exact sentences, and that the code literal and the classifier exist in exactly one place. |
 | `scripts/run-no-allow-cause-test.mjs` | Why an automatic lane did not allow a call, and the refusal line for the two automatic-refusal lanes. `noAllowCauseOf(card, ladder)` combines the card's standing-rule verdict with the host's rule-ladder evidence (only an own, strictly `true` property counts): allowed or denied by a rule gives no cause; a local ask rule gives `ask_rule`; an exhausted automatic-approval budget gives `budget`; an ask rule on the engine side (the card's class or origin) gives `ask_rule`; complete no-match evidence on a card a rule could clear gives `no_allow_rule`; everything else gives `other`, so no lane suggests adding an allow rule when that would not help. `policyDenyReason` mints the single line for the lane without an approval UI and for the dontAsk mode, one form per cause, naming only this one call and offering host-specific remedies (an allow-rule flag, a budget setting) only when the host supplies them. The guard covers the golden cases, pairwise ordering, strict reads, all eight sentences verbatim, unsafe tool names kept out of the sentence, and the single mint site. |
 | `scripts/run-approval-card-face-projection-test.mjs` | Three card positions added in 0.94.0. `boundGrants` (what approving a Skill card also grants: the skill name or null, the entries that then run without asking, and the entries refused for the rest of the run) is read once from the live approval frame through a strict reader; any malformed part drops the whole position, the copy is frozen, the digest is not put on the card, and the durable pending-row path never carries it; `approvalBoundGrantsDetail` gives one sentence with counts only, never the deployment text. `classifierNoRuling` (why the auto-mode classifier gave no ruling) is read by word shape as an open set from the live frame only; `classifierNoRulingDetail` has a phrase for each of the engine's words and a generic sentence otherwise. `priorCardsForCall` lists, for the same tool call, the earlier cards that asked about something else (a different class set, origin or mandate word), each with its approval id, origin, mandate word and, when answered here, whether it was allowed or denied; replays and parked twins of the same question do not count, both card paths share one bounded ledger, and `approvalPriorCardsDetail` says this is a separate check that needs its own answer. |
-| `scripts/run-hitl-words-readout-test.mjs` | The five wording and reader fixes of the HITL surface. (1) Seven user-facing failure sentences (pending-row fetch, approve and deny decisions on approval cards, the question-card decision, the parked-row wait, local rule parsing, background stop) carry the thrown value's own message and never the error class name — a thrown value with no usable message becomes one fixed sentence instead of a bare class name, `[object Object]` or `null`; the original thrown value goes only to the host log. (2) `resumeContextUnavailableContent` is a single English source (three ` · ` segments, with and without a run id, the retention seconds only when the engine gave them) and a plan-review decision that meets a 409 `conflict.resume_context_unavailable` says it. (3) The plan-review `permissionModeAfter` pre-check no longer mirrors the server's decision pairing: only the value set is checked locally, and the engine's `request.field_conflict` speaks for the rest. (4) `requestEngineCompact` reports how the fire settled through `onSettled` (accepted / refused / failed / offline), with a 409 `compact.not_running` also surfaced on the host surface. (5) Every exported `project*Capability` swallows a throwing getter into `undefined`. Pinned with throwing and class-named values at all seven sites, both plan-review entry points, the compact outcomes under 409 / 404-retry / 5xx / the 60 s injected deadline, and a per-projector throwing-proxy row plus a source scan. |
-| `scripts/run-session-read-redaction-capability-test.mjs` | The session-history redaction capability reader. From 7.120.0 the server's capabilities body carries a constant bit saying that model and tool output read back from a session's history is redacted by the same pass as the live frames; the reader keeps three states (the literal `true` is present, a missing key is an older worker — including the release that already redacted but did not report the bit — anything else is dropped as unreadable and reads as not observed), ignores keys that exist only on the prototype, reads the top-level key once and never throws, and its stored reading is discarded automatically when the engine's capability cache changes generation. One judge, `sessionReadRedactionAdmitted`, answers true only for a reading whose own data properties say present and `true` (accessors are never run). The bit stays true when the history route itself is absent, since it answers whether history is redacted, not whether it can be read; and the package source contains no version-number threshold for the same question. |
+| `scripts/run-hitl-words-readout-test.mjs` | The five wording and reader fixes of the HITL surface. (1) Seven user-facing failure sentences (pending-row fetch, approve and deny decisions on approval cards, the question-card decision, the parked-row wait, local rule parsing, background stop) carry the thrown value's own message and never the error class name — a thrown value with no usable message becomes one fixed sentence instead of a bare class name, `[object Object]` or `null`; the original thrown value goes only to the host log. (2) `resumeContextUnavailableContent` is a single English source (three ` · ` segments, with and without a run id, the retention seconds only when the engine gave them) and a plan-review decision that meets a 409 `conflict.resume_context_unavailable` says it. (3) The plan-review `permissionModeAfter` pre-check no longer mirrors the server's decision pairing: only the value set is checked locally, and the engine's `request.field_conflict` speaks for the rest. (4) `requestEngineCompact` reports how the fire settled through `onSettled` (accepted / refused / failed / offline), with a 409 `compact.not_running` also surfaced on the host surface. (5) Every exported `project*Capability` reads only own data descriptors: an accessor is never run and reads as `undefined`, and a throwing descriptor read is swallowed the same way. Pinned with throwing and class-named values at all seven sites, both plan-review entry points, the compact outcomes under 409 / 404-retry / 5xx / the 60 s injected deadline, and a per-projector throwing-proxy row plus a source scan. The decision exits treat hostile thrown values uniformly: a throwing message getter, a throwing toString, a revoked proxy, or throwing status, error-code or name getters all end in the thrown value's own message or the fixed fallback sentence, never in the getter's own error text, and the transient-failure classification reads each slot independently. |
+| `scripts/run-session-read-redaction-capability-test.mjs` | The session-history redaction capability reader. From 7.120.0 the server's capabilities body carries a constant bit saying that model and tool output read back from a session's history is redacted by the same pass as the live frames; the reader keeps three states (the literal `true` is present, a missing key is an older worker — including the release that already redacted but did not report the bit — anything else is dropped as unreadable and reads as not observed), ignores keys that exist only on the prototype, reads only own data properties (an accessor is never run and reads as unreadable) and never throws, and its stored reading is discarded automatically when the engine's capability cache changes generation. One judge, `sessionReadRedactionAdmitted`, answers true only for a reading whose own data properties say present and `true` (accessors are never run). The bit stays true when the history route itself is absent, since it answers whether history is redacted, not whether it can be read; and the package source contains no version-number threshold for the same question. |
 | `scripts/run-output-style-seat-test.mjs` | The per-turn output-style seat. From 7.121.0 the request accepts a top-level `outputStyle` object (`name` and an optional `turnReminder`) that the engine states once per attachment cycle, and the capabilities body carries a constant top-level bit for it; the reader keeps the same three states as its constant-bit siblings and does not confuse it with the nested bit for the style body. The live-defaults layer places the host's `outputStyleSeat` on the request only when that reading is admitted, copies it as given (an empty name or extra keys are passed through for the engine to refuse at prepare time with `config.output_style_invalid`, which the configuration-refusal table recognises), never overwrites a caller's own top-level value, never invents one, and leaves the style body seat unchanged; the field is registered in the live-defaults list, survives the unregistered-key check on both lanes, and is reported as belonging to another channel when passed to the request assembler. |
 | `scripts/run-request-uplift-test.mjs` | Six request-assembly and reader pieces shared by every client. The per-run declaration that nobody on an unattended run can answer approval requests is carried only on the unattended lane, only when the engine reports it accepts that field (an older engine rejects the whole request), with one plain sentence for each reason it was held back; the headless permission mode is always stated explicitly, falling back to `default` when neither flag nor environment sets it; a predicate tells apart the denial an engine makes when nobody is reachable and no queue can hold the request; a host-reported route problem (no gateway address, or the anthropic channel with no credentials) turns a model that has a name but cannot start into a stated configuration problem; a reader gives, per model, which thinking levels the engine accepts; attachments from MCP resources and the new not-attached reasons are written as data envelopes rather than as tool calls; and servers the client mounts itself state their own allow rule, audited separately from the user's rules, with the rule-source trust order exported once. |
 | `scripts/run-server-7122-pick-test.mjs` | Checks for the 7.122.0 server pickup. The new refusal for a leader run on a deployment whose autonomy is plan is recognised by its code alone and given one plain sentence; the parked-approval answer's new `message` is read only as an own non-empty string, never placed in the package's own sentence, across both the older and newer body shapes; the end-of-stream frame a draining server sends on read-only subscriptions is told apart from failures and from the pre-stream refusal with the same code, the reconnect wait comes from the same back-off as the pre-stream retry, the package's own stream readers stop treating that frame as a change or as an unknown frame, and a real published server is drained over its own endpoint to show the frame and the reconnect; and agent tool lists are checked for retired spellings and letter-case twins, entry by entry in agreement with the engine's own deny-list judge. |
@@ -502,6 +502,29 @@ guard still cross-checks the table by name).
 | `scripts/run-seat-wave2-test.mjs` | The seat capability table, the folder picker and the session folder, added in 0.94.0. Whether a renderer may offer an optional seat action is answered only by the capability table the host publishes (`seatCapability`: reported, not reported, or unreadable — reading own data properties only, never running a getter), never by calling the action and watching it fail; "not reported" is neither folded into "available" nor into "cannot", and the disabled sentence comes from one place (`seatCapabilityUnavailableDetail`). The subscription helper subscribes before it reads, so a renderer that subscribes after the host has published and then withdrawn a capability sees the withdrawn state on its first callback. `pickFolder` returns a non-empty path or `null` for a cancel, and the session record's `cwd` is the folder the session was started in, absent when none was chosen. The same gate pins `deviceFlowPollingPlan`: the device-code response's own positive finite numbers win, anything missing or malformed falls back to the contract numbers generated from the settings schema at build time (no number is written by hand), and the poll interval is never below one second |
 | `scripts/run-wiring-instructions-attribution-test.mjs` | Which project instruction file the engine actually loaded for a run, attributed to the client's own directory. The capabilities bit `callerCwd` gets its own reader (the same four ports as every capability reader; `callerCwdAdmitted` accepts only an explicit `true`), and a host that notes the bit inside its probe callback reads the new value in the same tick. The working directory a submitted request carried is recorded per run in the same in-process ledger and with the same frame-binding rule as the file-history opt-out (`legCwdOfBody`, `declaredLegCwd`, `legCwdOf`). `engineSelectedInstructionsFile` then returns the selected member only when the run's directory equals the client's current directory, the member is a plain relative path (no `..`, no absolute or drive form, no backslash, no control, bidi or invisible characters), and the host's real-path function confirms it stays inside that directory (a symlink pointing outside is refused). |
 | `scripts/run-layering-direction-test.mjs` | The package's layering table (`docs/LAYERING.md`) and its direction gate. Every source file sits on exactly one internal layer (wire readers and injected ports, projection, host-facing orchestration, the two entry points) and may import only downward, type-only imports included; the exemption table only shrinks. Every import that leaves the package must match a row of the cross-repository table and the kind that row allows: the engine package is read for types only and never at runtime, and the shipped declaration files reference only the packages listed in `dependencies` and `peerDependencies`. |
+| `scripts/run-panel-reduce-test.mjs` | The panel row reducer: one record per background agent, one alias index for every wire identity of it (fleet row id tail, transcript id, delegating card id), one generation rule, one place that constructs terminal facts and one place that emits a `terminal` row delta. `reducePanelFrame(state, frame)` is pure — no I/O, no clock read (the projection frame carries its own `nowMs`), no module state, the input state is never mutated, the same input always yields the same deltas and replaying a frame yields none. The delta set is closed (`upsert` / `usage_delta` / `terminal` / `remove`); `subscribeEnginePanelRows` replays a snapshot on subscribe. Twelve end-to-end sequences cover interleaved lanes, the first-published id, revival, reconnect/absence, late subscription, short-lived agents in both orders, historic rows, absent usage never filled with zero, the turn-end sweep scope, the transitional event channel versus the row channel, bounded eviction and snapshot fidelity. The generation criterion now lives in a dependency-free leaf module that both the panel record and the fleet ledger share; each layer keeps its own state (the panel's per-row watermark, the ledger's per-task watermark) but the comparison and the notion of a valid cycle number exist once. A data-flow scan keeps cycle numbers and watermark readings out of every comparison operand in the ledger (with planted violations to prove the scanner speaks, and no false alarm on presence checks or shape tests), and the numeric view is pinned against an independently written integer comparison over a grid that includes malformed values. |
+| `scripts/run-sdk-witness-compile-test.mjs` | The package source compiled against every SDK generation the peer range promises: each installed SDK at or above the declared peer floor (read from `package.json`, not hand-pinned in the guard) is pointed at through the compiler's `paths`, with the package's own `tsconfig.json` options, and `src` must produce zero diagnostics against each one. The ruler is proven to speak on every generation: the program covers every `src` file found by an independent directory walk, every SDK import in `src` resolves into the generation under test rather than the development dependency, and an in-memory planted call that omits the required `baseVersion` and a status word outside the closed union must both be reported while the corrected call stays silent. A probe tree installed on disk but missing from the witness list is red; an absent probe set is an honest skipped section, never green. |
+| `scripts/run-interactive-interrupt-emitter-test.mjs` | The interactive interrupt sender (`fireInteractiveInterrupt`): one bare `POST /v1/runs/:id/interrupt` through the client's runs resource, with the method called on its receiver, the session parameter sent only when non-empty and no idempotency key, so exactly one request goes out. Every outcome is a value, never a throw: a receipt yields `halted` with `turnCut` read only from an own boolean data property; a thrown error is classified by the same at-most-once judge the steer and cancel paths use, so a definite refusal becomes `refused` with its status and code, while a 5xx, a network failure or an abort becomes `transport` (no verdict). A missing resource yields `unarmed` with zero requests. The suite cross-checks every thrown sample against the shared judge, drives a real SDK client with a stubbed fetch and feeds the outcomes to `planInteractiveHalt`. |
+| `scripts/run-device-rebind-judgement-test.mjs` | The device rebind flow (`listDevices`, `readSessionBinding`, `rebindSession`, `reconcileRebindOutcome`, `rebindProvesRequestNotApplied`): each verb goes through the client's devices resource and returns a discriminated outcome, never a throw. Refusal codes are matched as status-and-code pairs from one closed table (device lane absent, run store absent, not found, active run, revision conflict, service token required); any other failure of the rebind write is classified by the shared at-most-once judge into `refused` or `unconfirmed`. A rebind success is accepted only when the echoed session, the target device and the revision step all hold, and the workspace carryover disclosure is carried only for the literal `none`. Response fields are read as own data properties. Includes a real SDK client with a stubbed fetch and a full grid of the read-back reconciliation. The eleven device row types from the SDK (`DeviceSummary`, `DeviceBinding`, and the rest) are re-exported as types from the package root, and the guard checks each one is identical to the SDK's. |
+| `scripts/run-engine-env-sourcing-test.mjs` | The local-engine environment group sourcing judge (`resolveEngineEnvSourcing`, `engineEnvView`): when a settings source wrote half of an all-or-nothing group and that half is ignored, the group is either skipped (its own switches are dropped) or, for safety groups and groups declared not skippable, left for the engine to refuse with an explanation. Group declarations, the ignore rule and the alias table come from the host; the module holds no retired-name table and reads no platform state. The suite runs a seeded differential test against a frozen transcription of the previous host-side loop, fixed cases for alias precedence, source precedence and case folding, and scans the built module for every retired environment name in the settings contract. |
+| `scripts/run-models-dropped-entries-projection-test.mjs` | The live model catalog's dropped-entries reader (`droppedModelEntriesOf`): which catalog entries the engine did not accept and why, read from `GET /v1/models` `droppedEntries`. An absent key (an older server) reads as `not_reported` — never as an empty list — while a present empty array is a real "nothing was dropped". Rows keep the server's order; a malformed row (non-integer index, empty reason code, bad identity, accessor fields, sparse holes) is counted as unreadable without dropping its siblings; unknown reason codes and additive keys pass through unchanged. Only own data properties are read, accessors never run, a throwing read never escapes, and the result is frozen. The exported row type is derived from the SDK's `models.list()` return type, so it compiles on the peer floor SDK as well. |
+| `scripts/run-model-catalog-read-projection-test.mjs` | The single reader of the `GET /v1/models` body (`readLiveModelCatalog`) and the readers built on the catalog it returns. A row is keyed by its `name` (the value a request submits and the `@` handle); a malformed row is counted and dropped without losing its siblings, a later row repeating a name is counted, and the result is a frozen deep copy. The default model is the server's answer read from `defaultModel` only (`liveDefaultModelOf`): two rows sharing an upstream id resolve by `defaultModel.name`, a default missing from the catalog reads as `not_listed`, and the first row is never offered as the default. Per-row readers answer whether a model may be `@`-mentioned (`atMentionOf`: an absent key means the deployment has no mention list, not "cannot be mentioned") and what line-compatibility declaration it carries (`modelCompatOf`); the accepted thinking levels stay with `effortLevelsOf`, never re-derived from the compatibility block. `liveModelRowsOf` gives one frozen view per row with every field taken from those same readers; a context window is carried only when it is a positive finite number, with no lower floor. The desktop seat's `listModels` returns this same catalog and its outbound gate shares the row rule. |
+| `scripts/run-seat-carriers-test.mjs` | The desktop seat's request carriers and the park bit. One inbound judge for `sendMessage` (`checkSendMessageRequest` / `isSendMessageRequest`): a closed key set, each key's vocabulary taken from the package's own tuple (permission modes, reasoning efforts, rewind modes, turn settings) and the attachment cap; the two subagent keys are judged (boolean, or `true` / `{ ttlMs?, max? }` with non-negative integers); an unregistered key, an accessor or a throwing read is refused with a key path. The rewind carrier `{ mode, targetMessageUuid }` uses the same tuple as `rewindSpecForMode`, and the request keys it is projected to equal what that projector emits. `SEAT_TURN_SETTINGS_ORIGINS` records, for every settings sub-key of the request builder, whether the renderer carries it or the host supplies it; host-supplied keys are refused on the seat. Optional seat positions — verbs and carrier keys alike — are answered only by the seat capability table (`rewind` added). `tool_permission_resolved` may carry `parked: true` only when its outcome is `expired`, minted from the closing frame through `approvalCompleteParked`, the same reader the frame router uses. |
+| `scripts/run-hooks-managed-policy-reading-test.mjs` | The optional `SettingsPort.managedPolicyReading()` that tells the hooks projection whether the managed policy layer could be read. Before it, only a read that threw counted as an unreadable governance layer, while a host's per-source reader usually does not throw: a managed file that cannot be read comes back as null and the projection treated that as no managed policy, so user, project, local and flag hooks went out to the engine exactly when an administrator's restriction had disappeared. The reading has three words. `absent` (readable, nothing configured) and `in-effect` (readable, has content) project exactly as before — the governance gates in the managed document still decide — while `unreadable` closes the whole leg: no hooks from any source and no session-level goal hook are sent, and enabled plugins are listed as excluded for an unreadable governance layer. The change only tightens: for the three readable answers (reader missing, `absent`, `in-effect`) the projection is byte-identical on the same managed document, and `absent` never overrides a document that says hooks are disabled. A reader that throws, or answers anything that is not one of the three words (including a different letter case, an empty object, or a property where a method is expected), counts as `unreadable`; only a missing member means the host did not wire the port, which keeps today's behaviour and logs nothing. |
+| `scripts/run-park-expiry-judgements-test.mjs` | The judgements the approval flow needs once a streamed approval's window has closed and the engine has moved the call into its durable queue instead of refusing it. `classifyApprovalCompleteFrame` reads a closing frame as one of three outcomes (moved to the queue, no queue evidence so retract the card, decided by a person) or none for anything it does not recognise; only a literal `true` counts as queue evidence, and a new outcome word is never mistaken for an expiry. `parkRowDeadlineState` reads a durable row's `deadline` into none, pending or past, treating a missing, null or absurd value as no deadline rather than clamping it. `lateDecideRoute` picks the endpoint for a late decision (the durable decide endpoint only when the window had closed and the row was actually found). `runReapedByDeadline` reads the run record and answers true only for the reaped-by-deadline error code; anything unreadable is false, so an unreadable record is never reported as a step that was not run. The guard carries the terminal client's offline cases over one by one and runs a frozen copy of that client's original implementations side by side over more than fifteen hundred inputs; the registered differences, both on the conservative side, are that a host client whose property read throws is answered with false instead of a rejection, and that a park bit sitting on the prototype chain or behind an accessor is not read as moved-to-the-queue (it is read through the same single reader the desktop seat uses, which only accepts own data properties). |
+| `scripts/run-usage-analytics-projection-test.mjs` | Reads of the engine's usage analytics (`GET /v1/usage/summary`, `/series` and `/breakdown`) and the card view built from them. Each reader never throws and answers ok, unavailable (the engine says it has no such face: no run store, no such route, or the feature is switched off) or failed (any other read failure, which is never reported as the backend lacking the feature). A missing number is simply absent from the view, never zero, and a total counts as exact only when the engine says `estimated: false`: otherwise it is an estimate and the view says so — a lower bound for money, an approximation for tokens, each with a shared prefix for the host to render. The card view has three states: available, estimated, and unavailable with one sentence for a backend that has no analytics and a different one for a read that failed; an unavailable card carries no numbers at all. The heat map and the active-day count exist only when the day series was both read and read completely, so a failed read, a series with unreadable points or invalid dates, or one the engine reports as truncated is never drawn as a grid of zero-activity days; the peak hour and the favourite model follow the same completeness rule, and a failed per-model breakdown is different from a window with no model usage. The guard also runs a real SDK client against a stubbed fetch and checks the four requests it makes. |
+| `scripts/run-engine-caps-single-source-test.mjs` | Engine capability reads have one store: the shared capabilities snapshot. The subagent tail, subagent report and manual-compact paths read it (probing once through it when nothing has landed, with a bounded wait) instead of keeping private per-address tables, so a capability follows `invalidateEngineCaps`; a connection without an identity key is probed once and nothing is stored. The suite checks the internal read path, the three migrated entry points against a host-kicked snapshot and across invalidation, an AST scan that every capabilities call in the source is a probe handed to the shared cache (two listed exceptions), and the retirement of the store-based two-generation readings and of the second argument of `toolsRunHereFromExecutionLane`. |
+| `scripts/run-mcp-engine-leg-connection-test.mjs` | The engine-hosted MCP connection reader (`mcpEngineLegConnectionOf`): whether a server is connected on the engine's leg is read only from that leg's `wiring_manifest.mcp[]` status, never from the tool roster or a client's own connection. The suite checks the three states and their causes (connected with an optional tool count, not connected because it failed, was skipped or is not listed, and unknown because the section was not received, the name matches several rows, the status is unreadable or the word is not recognised), that the row lookup is shared with the liveness reader, that the liveness record is not read, and the same answers after the package's own section projector, including rows the projector drops because their status is unreadable (counted by name on the projected array, invisible to JSON and key enumeration). |
+| `scripts/run-model-switch-fallback-test.mjs` | The model-switch judgments (`modelSwitchProbeOutcomeOf`, `runModelSwitchProbe`, `createModelSwitchFallbackLedger`): the probe outcome for each lane (allowed without probing on a mock lane, not switched when the probe is not wired, threw, was missing or rejected the name, with the probe's own words as the reason), the orchestration that only calls the validator when it is needed, and a per-session fallback ledger that names the last model that answered when the newly selected one fails its first turn with an empty stream. The suite checks every arm, the negative cases (other error codes, other sessions, other models, a model already proven), that judging never changes the ledger and that a stale sequence number cannot release a newer switch. |
+| `scripts/run-approval-cross-leg-holder-test.mjs` | One holder per approval across the live card and the durable paths. While the package holds the live card for an approval (the card is on screen or its answer is being sent), a durable path for the same approval (the package's own parked-row decision path, or a host's own durable route that asks `joinApprovalHolder` first) presents no second card and sends no decision; once the holder lets go, an approval it settled is joined (the holder's outcome marked `joinedInFlight: true`), and one it did not settle (the answer was not delivered, the row is being parked or was parked, the host withdrew the card) is taken over by the durable path, never while the live card is still on screen. Identity is the approval id when the caller has one; otherwise the same run and the same gated call (and the same session when both sides know it), matched only while the live card still waits for an answer, and any component that cannot be read means no match. Cells cover the host timing that produced the second card (an answer after more than the 750 ms grace window: zero second card, zero durable decision, the next approval's card shows at once), the negative controls (nobody holds it; the live answer fails with 404 and the durable card is requested only after the live card was answered), unreadable or different identities, aborts from either side (one interruption deny), the parked-row path joining an allow and a deny, a second approval of the same call while the first answer is in flight, a parked close-out handing over to the parked row, and the scope: a host that picks the route by frame type with no timed fallback is not on this path |
+| `scripts/run-engine-cap-snapshot-readers-test.mjs` | Capability readers derive their reading only from the capabilities cache's landed response, never from a second table: the same reading comes back without any tee call, the retired tee ports write nothing (a roster of all 29 names is pinned, equal to the root entry's exports and to the reader table), and before a landing, after an invalidation, after a failed probe, or for a malformed or non-object body the reading is "unobserved". A reading is memoised per landing and frozen, a second landing of the very same body re-projects, and re-entering a reader while it is projecting answers unobserved without being remembered. Every projection (all 29, plus the attachment-seat placeholder and the shared store-seat reader) reads only own data descriptors: a recording proxy sees zero `get` and `has` traps, an accessor at the top level or at any nested position that would return a valid value is never run and reads as malformed, a key that exists only on the prototype is not reported, and a throwing descriptor read is malformed rather than escaping. The seat gates follow the same law for every validated position of every checker (an accessor returning a valid value is refused without running, a recording proxy sees no `get` trap, a read that throws is a refusal and never a throw), and a payload with an accessor anywhere in its tree — validated or passed through — or a validated position that is not enumerable is refused with a dotted path, because the structured clone a host performs afterwards would run the accessor and rewrite a position that had already passed. Finally no source file in the package calls a `noteEngineCapsFor*` function, the factory's `note` has an empty body, and the judge that decides whether the "edit and rewind" hint is spoken is wired when the module loads rather than by any tee call (checked in a fresh process that only lands bodies and never tees). |
+| `scripts/run-server-7124-pick-test.mjs` | Checks for the 7.124.0 server pickup. The reason an approval ended without a human answer is read from whichever body key the server sent — the new one when it is present, the older one only when it is absent — against one closed list taken from the SDK, grouped by end state, so both generations of server give the same reading; the live approval card's ending reason is read as a closed list and its absence is never filled in; the decision endpoint's conflict refusals are split into a binding change, a damaged row, an unsupported stored format and an invalid answer, and an invalid answer is no longer reported as "the pending action changed" on either server generation; the two version fields on the unsupported-format refusal are read together or not at all and change the way out only when the stored work is newer; interrupting a question that is no longer pending (already settled, parked or gone) no longer shows the "session may stay locked" warning, while real failures still do; the run-retired code list matches the SDK's; the SDK itself ends a subscription on its final frame; the engine's registered worker name is projected only when sent. |
+| `scripts/run-run-events-drain-reopen-test.mjs` | A draining server ends each read-only subscription with a final frame, and from SDK 19 the run-event stream simply ends there instead of reconnecting. The two run-event legs (the stream that resumes a run after an approval decision, and the headless lane's continuation) now reopen the same run from the last event they saw, after the same back-off as the other subscriptions (1, 2 and 4 seconds, three times), without sending a second decision, without counting the drain as a connection failure and without showing the drain frame; the run carries on to its normal end. The drain frame is never handed on. When the back-off is used up or the run handle is unknown, the leg ends the way a dropped stream does (the decision leg ends cleanly with no final frame and without inventing a failure, so the host's own handling of a dropped stream takes over; the headless leg first looks up the run's stored result, then falls back to its failure path); a caller that has stopped gets the usual interrupt exit. A thrown error is still a lost connection and takes the failure path, and an SDK that reconnects by itself never gets a second stream. All four subscription readers share one drain bookkeeping helper, checked by a syntax-tree roster. |
+| `scripts/run-skill-tool-lists-projection-test.mjs` | Two tool lists on a request skill, `allowedTools` and `disallowedTools`, are only put on the request when the engine reports that it reads them. `skillCommandsToSpecs` takes the capability reading as an option, asks the one capability judge, and sends the lists only on a clear yes; with no option, an unobserved, missing or negative reading, or a value that is not a reading, the request is exactly what earlier versions built, so an older engine never sees the keys (it ignores them on a top-level skill and rejects a delegated agent's skill outright). When they are sent they go as given: no trimming, de-duplication, reordering or truncation, an empty list stays an empty list, and the caller's array is copied rather than shared. The only local check is the shape: each list must be an array of strings or absent (an array that cannot state a valid length, or a read that throws, is refused too), otherwise the call is refused with a sentence naming the skill and the list (never echoing entries or the thrown value), and once the option is passed that check runs whether or not the lists would be sent; without the option the lists are not read at all. Entry-level limits (the per-list cap, deny entries the engine reads as nothing, empty strings) are left to the server, which refuses loudly; the suite runs the server's own admission functions for every fixture generation from 7.116.0 on and shows the package output accepted, the over-cap list still carried in full and refused with the server's own sentence, and the unreadable deny entry still carried and named by the server's judge, and for older generations shows the output carrying no list keys and accepted, while a delegated agent's skill with the keys is rejected. The same output fed to the delegated-agent projection gives the same shape. A source scan pins that there is exactly one reader of the capability and no copy of the server's limits. |
+| `scripts/run-engine-target-facts-projection-test.mjs` | The engine-target fact judgments (`engineCredentialRequirementOf`, `storePostureHeadWord`). The first reads the answer to one credential-free probe of an engine — an explicit success, an HTTP error status, or no answer at all — and returns `open`, `requires` or `undetermined`: only an explicit success is `open`, only 401 and 403 are `requires`, and everything else (any other status, a transport failure, contradictory or malformed answers, inherited or accessor keys, hostile proxies) is `undetermined`; it never throws and holds no host state. The second returns the narrow-panel lead word for a store that does not persist its state, only when the persistence judgment says so (never for a durable, unreported or unobserved store), and it is the same constant the full doctor sentence uses, whose text is pinned byte for byte. The suite sweeps every HTTP status, cross-checks a generated pool of answers against an independently written reference, and pins that the lead word and the full sentence share one source. |
+| `scripts/run-session-id-path-segment-test.mjs` | The session-id path-segment judgment (`isSessionIdPathSegment`). A session id is spliced into `/v1/sessions/<id>/…`; URI encoding leaves `.` and `..` untouched and URL resolution swallows them as dot segments, which re-aims a request at a different route, so the id is usable only when it is a non-empty string, not a dot segment, and encodable. The suite cross-checks the judgment against a generated pool resolved by the platform URL parser, drives every call site that hands a session id to an engine route (memory erase, memory status, MCP panel and reconnect, background stop and agent registry read, delegated prompt) and checks that dot segments send nothing while a valid id still goes out, and keeps a syntax-tree roster so a new call site that skips the judgment turns the gate red. |
+| `scripts/run-rewind-engine-action-test.mjs` | The two pure judgements the terminal client used to keep for its rewind command. The dispatch gate says whether, right now, the engine has explicitly said that a given rewind mode cannot be done: each of the three modes (rewind with code restore, rewind the conversation only, restore code only) reads the same availability judges the package already ships for the menu, only an explicit off refuses, and a missing, unreadable or unobserved reading never refuses. The action judgement takes the chosen option, the target position the host has already resolved (session root, after the root, or undecidable, with the engine entry handle if there is one) and the capability reading, and answers whether to arm the one-shot rewind keys, to reopen the engine session (the two modes that carry a resume point, chosen at the session root, which the server will not rewind past) or to do nothing, with a reason; an undecidable position never acts in either direction, and the code-only mode looks only at the handle. The preview used by the confirmation screen is the same answer's kind. The guard checks both judgements against hand-written truth tables over every bit combination and every target shape, that the dispatch gate and the menu can never disagree, that the default reading follows the current engine target, that hostile inputs never throw and never act, and that the module keeps no mode list or resume-point list of its own. A live comparison against the terminal client's own copies runs in the layering guard. |
+| `scripts/run-refused-park-family-test.mjs` | The refused-resume family (`refusedParkFromError`, `refusedParkContent`, `refusedParkBlockedContent`): three 409 codes whose refusal comes back every time the same decision is sent again (the stored park is not supported by this engine version, its resource ledger carries a retired wall-clock allowance, or the engine's gate refuses its stored task configuration or permission-mode seed). The suite checks the reader on SDK error objects and on flattened failure facts (status policy, field token, still-pending reading, getters that throw), compares every sentence byte for byte against a frozen copy of the terminal's previous implementation over each combination of code, field, pending state, lane and run id, pins the deliberate differences one by one, and checks that the sentences are correct per lane, never print an unknown code, pass the package hygiene list and do not trip the already-resolved word scan. |
 
 Each suite carries a floor that only moves up — a refactor that stops executing a group of
 assertions is a failure, not a quieter pass. Guards anchor on the **installed artefact's content**