@sema-agent/core 5.39.0 → 5.41.0

package/CHANGELOG.md

@@ -1,5 +1,144 @@
 # Changelog
 
+## 5.41.0 — 2026-08-18
+
+No BREAKING changes. Three behavioral narrowings disclosed below (garbage values on the three
+tool-face seats now refuse loudly; the strict sidecar family's journal read and lock fence both
+join the fail-closed law).
+
+### Added
+
+- backlog #259 — user steers stranded at agent_end are ANNOUNCED, never silently dropped behind
+  their "queued" receipts: the terminal sweep hands ENGINE notes back for per-session redelivery,
+  but USER inputs have no redelivery semantics (a steer aimed at a finished run must not fire at
+  the next one) — their remnant counts now ride a dedicated harness sink, the `settled` frame
+  (additive `undrainedSteerCount`/`undrainedFollowUpCount`), and a new `EngineNotice` family
+  `task.user_steer_undrained`. The stranding window is a narrow race past the loop's final queue
+  check, which is exactly why it gets a loud terminal account.
+- backlog #251 — the mixed-version window of the design/186 ledger v2 is machine-detected:
+  recovering a v1-format journal over a v2 disk ledger (a shape only a pre-v2 process mints)
+  announces the old-writer-alive warning on the external-findings channel; recovery is unchanged.
+
+### Fixed
+
+- backlog #297 (ruled: narrow) — the three tool-face seats (`excludeTools` / `deferTools` /
+  `alwaysLoadTools`) get the fourth seat's value door: a garbage value (a bare string, a truthy
+  non-iterable, a non-string entry) refuses `config.tool_face_control` instead of stringifying
+  into a never-matching list, throwing an uncoded TypeError, or a falsy value silently reading as
+  absent. **Narrowing**: a deployment whose misspelled seat previously "worked" now hears it.
+- backlog #299 (ruled: fix alone) — the auto-mode classifier's action block bounds tool args at
+  48,000 chars (its own subject-level cap, deliberately above the delegation-review lanes'
+  12,000-per-field sampled payloads so their coverage notes ride through uncut): it was the one
+  unbounded string in the classifier's prompt, so a single long tool argument (a SendMessage body
+  needs nothing crafted, just length) could blow the classifier's context and trip the session
+  breaker out of auto mode.
+- backlog #307 (also filed as #256) — the journal-aware strict sidecar read joins the ENOENT-only
+  absence law: an EACCES/EIO journal read refused `ControlPlaneCorruptError` instead of silently
+  serving the pre-transaction state while a committed next state sat unreadable beside it — the
+  one fail-open left in the strict-read family. **Narrowing**: a deployment whose journal reads
+  were failing now hears it instead of silently reading stale.
+- backlog #255 — the sidecar lock's commit fence splits by family: the fail-closed LEDGER family
+  (challenge/lineage locked updates and the rebuild path) aborts on an UNPROVABLE owner
+  (absent/unreadable owner file) — "cannot disprove a steal" is not a fence; the calibration
+  family keeps its documented pre-token leniency, and a FOREIGN token aborts both as before.
+  **Narrowing**: within the ledger family only.
+- CHANGELOG erratum for 5.40.0 (recorded in that entry): the header counts THREE narrowings
+  (the defer alignment was the missing third) and the release date corrects to 08-17.
+
+## 5.40.0 — 2026-08-17
+
+No BREAKING changes. Three behavioral narrowings disclosed below (the requestedCwd × workspace-restore
+combination now refuses loudly; Grep's ripgrep leg withholds alias-spelled deny-listed paths it
+previously returned; `deferMode: "auto"` never sweeps engine built-ins — a small-window deployment
+whose built-ins were previously deferred now carries their schemas inline, trading context bytes for
+the first-use activation error the sweep caused).
+
+<!-- Post-release erratum (2026-08-17, server review [4351]): the header originally said "Two
+behavioral narrowings" and dated the entry 2026-08-18 (pre-written); the defer narrowing was listed
+under Fixed but missing from this header, and the release actually shipped late on 08-17. -->
+
+### Added
+
+- design/178 v2-c — governed memory export/import: `MemoryEngine.exportMemoryScopes` /
+  `importMemoryBundle` over a self-describing bundle (`MemoryExportBundle`: entries + governance
+  rows — unresolved challenges, durable pollution markers, committed lineage, custody evidence —
+  + residuals + a recomputable integrity section). Import is two-phase: entries land under a
+  synthetic lineage LATCH (withheld from every model-visible read face) and only a durable,
+  fully-validated completion receipt releases them; a crash mid-governance converges by re-running
+  the SAME bundle import (idempotent). Tenant slicing is per-artifact three-way (refusal /
+  field redaction with an import-visible provenance stamp / global-residual); a `complete:false`
+  export refuses to mint a package at all. Integrity detects accidents, never forgery (self-attested
+  hashes; authenticity is the transport's job — the multi-tenant service form gates imports behind
+  its authenticated endpoint). New exports: `computeMemoryBundleHash` + `MemoryExportBundle` /
+  `MemoryImportReport` / `MemoryExportSnapshot` / `MemoryBundleImportPlan` / `BundleChallengeRow` /
+  `BundleLineageRow` / `BundlePollutedSession` types. New coded errors: `memory.export_incomplete`,
+  `memory.import_rejected` (input-shape errors use the existing `config.memory_export_request` /
+  `config.memory_import_request` family). Free-string fields (pollution/challenge reasons, custody
+  `req`, session ids) export VERBATIM; the bundle's `doc` lines disclose that sub-scope tenant
+  isolation does not extend into those literals.
+- backlog #243 — new coded refusal `checkpoint.cwd_conflicts_restore` (see Fixed below).
+
+### Fixed
+
+- backlog #243 — `RunInternals.requestedCwd` combined with a checkpoint workspace restore is now
+  REFUSED loudly instead of half-working: the restored workspace's mountPath is authoritative for
+  the task root, so the cwd was semantically ignored at best — and verifying it was prepare's FIRST
+  env RPC, which a lazy-connect remote adapter answers by connecting a fresh EMPTY sandbox before
+  `resumeVM` (silently running the resumed task on an empty workspace), or the check itself
+  deadlocked the approval on a phantom mismatch (`env_failed` reopen retrying deterministically).
+  Two gates: the resume ladder refuses PRE-CAS (`checkpoint.cwd_conflicts_restore` — the checkpoint
+  stays pending; re-resume without the cwd succeeds), and prepare carries a depth twin for direct
+  `runTaskStream(spec, resume, internals)` callers (`config.cwd_conflicts_restore`, refused before
+  the env factory mints anything). **Narrowing**: on eager adapters this combination previously
+  "worked"; it now refuses on every adapter. Core's own delegation drivers never mint the pair.
+- backlog #303 — Grep's ripgrep leg gains a deny TRIPWIRE: rg's exclusion globs are spelled from
+  the pattern text and cannot express the win32 component-alias family (`.aws.` — trailing
+  dots/spaces the win32 namespace strips), so a directory planted under an alias spelling slipped
+  past the globs and its contents reached Grep results whenever ripgrep was installed, while the JS
+  leg correctly pruned it. Every rg output line is now judged with the same two-view matchPath the
+  JS walker prunes with; ANY deny hit — or any line the record format cannot account for (e.g. a
+  newline inside a path component splitting one rg record across physical lines) — abandons the rg
+  pass entirely and the JS scanner answers, pruning with the authoritative judge and declaring the
+  abandonment on both the text and structured surfaces. **Narrowing**: results no longer differ by
+  whether ripgrep is installed; rg remains the fast path when no guarded entry is involved
+  (`#306` tracks the structured `--json` terminal form).
+- backlog #298 — a THROWING deployment-supplied `breakerOpen` on the auto-mode decider reads as
+  not-open: the peer-referral ask is still minted (one extra question), never a silent allow.
+- backlog #300 — `WRITE_PROTECTED_DEFAULT_TABLE` and every resolved table are deep-frozen: an
+  admin-face in-place mutate (splice / `length = 0`) throws at the mutation site instead of
+  silently weakening the process-wide protection floor.
+- backlog #301 — the CC settings-import preview guards the settings ROOT like it already guarded
+  the `permissions` slot: a valid-JSON document that is a string/number/array reports the layer as
+  UNREAD (`skipped`) instead of two clean zeros; a `null` document remains the readable
+  "nothing written".
+- 5.40 pre-release rescan (memory-engine import/export hardening, all pinned):
+  - the import's retroactive pollution sweep mints its challenge events under a TOP-LEVEL
+    `imp-sweep:` prefix, structurally disjoint from the carried-challenge leg's
+    `imp:<hash>:<source eventId>` — a shared key with a stale frozen `challengedRev` (a
+    crash-window promote on the source drifting lineage `lastRev`) aborted the read-back verifier
+    AFTER the latch staged, and the documented re-import convergence reproduced the abort forever;
+  - BOTH export faces refuse over the durable chain-degradation marker
+    (`transfers.chain-degraded.json`): a healed (quarantined + truncated) evidence chain parses
+    clean, but the store has declared the lost rows unprovable, and a bundle minted over it
+    testified to a deletion history the source knew was incomplete (an erased id could resurrect on
+    the destination); the composite throws `memory.export_incomplete`, `governanceExport` answers
+    `complete:false` with the reason;
+  - the bundle-import completion receipt validates EVERY `MemoryImportReport` member — an 8-of-15
+    husk could release a standing latch and hand back a report whose missing arrays throw on first
+    touch.
+- `deferMode: "auto"` never sweeps engine BUILT-INS (align-with-CC ruling): the pressure valve's
+  candidates narrow to caller-supplied specs — CC's own `isDeferredTool` defers MCP as a family
+  plus individually self-declared tools and has no pressure valve at all, so built-ins like
+  Grep/Glob/Read stay inline unconditionally; sweeping them made every small-window model pay a
+  first-use activation error on tools its training treats as always present. An explicit
+  `deferTools` naming a built-in still honors the deployment's stated intent.
+- Out-of-root refusals no longer print a duplicated root: a directory declared through two doors
+  (`additionalDirectories` + the advertised scratchpad) deduplicates by canonical value in the
+  disclosure list (the judgment was already idempotent).
+- backlog #309 — the out-of-root escape hint names the `readFace: "open"` knob beside the
+  per-directory seats; the unknown-skill refusal states that foreign skill directories are not
+  loaded and points at importing the content instead (signpost ruling).
+
 ## 5.39.0 — 2026-08-17
 
 No BREAKING changes. Two behavioral narrowings disclosed below (write-protection default table;
@@ -63,7 +202,7 @@ baseline slots refuse at mount instead of silently mounting ungoverned).
   spelling was already collapsed by the filesystem itself — a genuinely-resolved POSIX target keeps
   its own reading and a real `.aws.` directory stays readable and writable. The arms that resolve
   NOTHING do not report it and keep both views: UNC keys (minted with zero probes for liveness) and
-  not-yet-existing targets (whose tail is rejoined verbatim) — which is where the planting vector
+  not-yet-existing targets (whose tail is rejoined verbatim) — which is where the pre-creation path
   lives. Unreported is the fail-closed default, so a caller that forgets to pass it loses precision,
   never protection.
   **Residual cost**: a target the filesystem has NOT resolved still over-matches — writing a
@@ -902,7 +1041,7 @@ breaking change (rows without the new gate bit keep their historic stamps).
   governed row (org origin) is refused PRE-CAS on a worker with no `permissionRuleOrg` wiring (the
   human decision stays unspent, redeemable on an org-wired worker; a post-CAS belt remains as
   defense in depth); a v8 row whose bit was stripped in storage, and a sub-v8 row carrying a bit no
-  release minted, are both refused pre-CAS (corruption/forgery guards).
+  release minted, are both refused pre-CAS (corruption/fabrication guards).
 - **`run_in_background: true` is judged by the backgrounding doctrine (#125, P0).** On a
   `shellGate:"classify"` deployment, `bashReversibilityProbe` used to read only the command text, so
   `ls` + the parameter form auto-admitted exactly what `ls &` asks for. The parameter spelling now
@@ -1636,7 +1775,7 @@ mysterious runtime.
   claiming the row is still pending. Terminal-state refusals name the observed status without
   coercing store-supplied values.
 - The approval lane's `reason` is validated as plain text at capture; malformed decisions and
-  hostile text shapes get typed refusals instead of raising inside the refusal path.
+  unexpected text shapes get typed refusals instead of raising inside the refusal path.
 - Orchestration guidance, the TaskOutput tool card, and the selective-recall affordance stop
   teaching retired tool names: the workflow tool is named by its wire name, and
   `composeSelectiveBody` accepts an optional caller-supplied `recallToolName` (additive).
@@ -2178,7 +2317,7 @@ live `onAsk` is wired — drop the approver (or arm `forceDurableGate`) to test
 
 - **Memory recall: an unknown write time is incomparable, not 1970.** A note whose `ts` failed to parse used to sort as epoch 0 — it lost every `[[name]]` collision (permanently unreachable through links) and rendered a fabricated age ("written 678 months ago"). `MemoryNoteHeader.timestampMissing` (new, optional) is minted at the single parse point; the one recency comparison treats missing as a tie (the incumbent keeps the name — an unknown-time note neither sinks nor poses as newest), and the render face says "write time unknown — verify it's still current". A literal 1970 timestamp string stays a real time (the numeric value is never the signal; negative pins lock this). Bonus: a wire row with NO `ts` key used to throw and kill the whole scope's manifest build. **Consumer flip**: a probe pinning the fabricated-age wording or the sink-to-bottom collision outcome reds.
 - **Three awaited rescue timers are REF'd + cleared-on-settle** (subagent observer drain, teacher abortFallback, WebFetch deadline): unref'd, the process could exit before the rescue fired when the wedged arm held no live handles — the settle never completed. The bg notify drain window keeps its unref by design (a background lane must not pin the process).
-- **A harness-side throw around ONE tool call becomes that call's disclosed error result — the batch survives.** Worker exceptions that are not cancellation (a rejected `tool_execution_update` delivery, a sink failure) used to reject the whole parallel batch, orphan the sibling calls, and on the worst path rewrite a COMMITTED side effect into `[tool execution harness failed: …]` — inviting the model to re-run an already-effective write. Progress deliveries now settle at one point; a delivery failure on an executed tool keeps `content`/`isError` untouched and rides `details.toolProgressDeliveryFailed` (total by construction: hostile accessors or exotic `details` decline the annotation, never the outcome; `afterToolCall` cannot erase it). Cancellation — including one wrapped in the harness's own `AgentHarnessError` (`cause`-chain walk) — stays a hard batch failure on every lane, and the in-stream lane closes admission immediately. The sequential lane shares the same disclosure band. **Consumer flips**: probes pinning the old `[tool execution harness failed]` rewrite or `isError:true` on a delivery-failed success red — re-pin on the new shape plus a retired-dialect tripwire.
+- **A harness-side throw around ONE tool call becomes that call's disclosed error result — the batch survives.** Worker exceptions that are not cancellation (a rejected `tool_execution_update` delivery, a sink failure) used to reject the whole parallel batch, orphan the sibling calls, and on the worst path rewrite a COMMITTED side effect into `[tool execution harness failed: …]` — inviting the model to re-run an already-effective write. Progress deliveries now settle at one point; a delivery failure on an executed tool keeps `content`/`isError` untouched and rides `details.toolProgressDeliveryFailed` (total by construction: throwing accessors or exotic `details` decline the annotation, never the outcome; `afterToolCall` cannot erase it). Cancellation — including one wrapped in the harness's own `AgentHarnessError` (`cause`-chain walk) — stays a hard batch failure on every lane, and the in-stream lane closes admission immediately. The sequential lane shares the same disclosure band. **Consumer flips**: probes pinning the old `[tool execution harness failed]` rewrite or `isError:true` on a delivery-failed success red — re-pin on the new shape plus a retired-dialect tripwire.
 - **openai lane: an evidence-free `tool_calls` fragment no longer fabricates an unnamed-call disclosure.** The evidence gate sits at the finalize disclosure point (slot assignment also routes id-less continuations, so gating there corrupted two real calls); the stream's own verdict wins — `finish_reason:"tool_calls"` with zero survivors, or an empty slot beside real calls, discloses loudly; an isolated terminal fragment with `stop`/`length` stays silent.
 - **A throwing tool keeps its machine-readable marks on BOTH legs.** Two strip points, one class: the `defineTool` adapter rewrapped a ToolSpec throw as a bare `Error` (marks died even LIVE — a ToolSpec throw carrying `details.code` got no `errorCode` lift while a native tool's identical error did), and the durable-resume batch's catch stringified the message alone (breaking the live/resume frame-parity contract). Both carry `details`/`errorKind` through now; message text unchanged.
 - **Cache-break attribution: an aborted / usage-missing row is "unknown", not a cold cache.** Feeding it to the design/31 detector minted a false `server-or-ttl` finding and poisoned the warm baseline for the next real turn. Such rows are skipped; a genuine cacheRead collapse still fires.
@@ -2708,7 +2847,7 @@ _Emergency direct-publish path (CI publishing resumes 8/1); pre-tag `gate:blackb
 **API retry chain (user-reported, rate-limited third-party providers)**
 
 - `maxRetries` defaults to 10 (was 2), configurable via explicit brain options (highest priority, uncapped) or `SEMA_MAX_RETRIES` (clamped to [0,15]); `retryDelayMs` base is now 500.
-- Backoff is increasing-with-light-jitter — `min(500·2^(n-1), 32000) + rand·0.25·backoff` — replacing full-jitter (attempt N could previously wait less than attempt 1). `Retry-After` keeps its existing max-composition semantics; `anthropic-ratelimit-unified-reset` is now honored the same way (header values clamped to 60s against hostile headers).
+- Backoff is increasing-with-light-jitter — `min(500·2^(n-1), 32000) + rand·0.25·backoff` — replacing full-jitter (attempt N could previously wait less than attempt 1). `Retry-After` keeps its existing max-composition semantics; `anthropic-ratelimit-unified-reset` is now honored the same way (header values clamped to 60s against out-of-range headers).
 - Retry liveness reaches the task event stream: `status` frames carry `attempt`/`maxRetries`/`retryInMs`, waits longer than 30s re-announce per slice, recovery emits a `recovered` terminal and exhaustion a `gave_up` terminal (phases added to `BrainStatusPhase`), and `circuit_open` carries the remaining cooldown. Clients can finally clear a stale error row the moment the stream recovers.
 
 **Security (read-only bash boundary — bypass closure, disclosure correction for 2.8.0)**
@@ -3486,7 +3625,7 @@ Carried over from 1.410.0's open list: **RB-95** (the compaction cut-point's bac
 
 ## 1.410.0 (2026-07-25)
 
-The divergent-attack round: four more probe agents, again forbidden to touch source, produced 44 reproducible defects across the approval chain, the session tree, the tool surface and the agent-team stores. 22 are fixed here, 2 are recorded as open with their reds resident. One of them was mine — 1.409.0's own fix introduced it, and it is the most serious thing in this release.
+The divergent-probe round: four more probe agents, again forbidden to touch source, produced 44 reproducible defects across the approval chain, the session tree, the tool surface and the agent-team stores. 22 are fixed here, 2 are recorded as open with their reds resident. One of them was mine — 1.409.0's own fix introduced it, and it is the most serious thing in this release.
 
 **Note on 1.409.0**: its tag was pushed but CI went red on `flaky-guard` and it never reached npm. The cause was not the release content — the guard's stale-allowlist check flagged a declaration for an untracked `*.local.test.ts` probe, a file that exists on a developer's machine and, by design, not in CI. The check now skips entries whose file is absent from the scan. Everything in 1.409.0 ships here.
 
@@ -3513,7 +3652,7 @@ A multi-dimensional red hunt (five independent probe agents + a first-party pass
 - **fix (RB-64): the anti-re-vote guard was nested entirely inside the `policy_ask` arm**, so a `plan_review` checkpoint reopened after an `env_failed` retry accepted a decision contradicting the one the human actually made — and `winnerFromOutcome` recorded no winner for that gate, leaving nothing to replay against. Winners are now recorded for the review gates (with a synthetic `gate:plan_review` sentinel that can never collide with a real toolCallId) and the guard is gate-agnostic.
 - **fix (RB-68): `durableApproval.ttlMs` reached `Date.now() + ttlMs` unvalidated** — `Number(unset_env_var)` produced `deadline = NaN`, and `NaN <= cutoff` is always false, so a checkpoint that carries a TTL and looks bounded could never be reaped by any cutoff. Fixed in both layers: mints normalize a non-finite/non-positive TTL to the documented default (a deliberate `undefined` still means unbounded), and `reap` treats an uncomparable deadline as reapable, since checkpoint rows are also written by deployment-side stores and older versions.
 - **fix (RB-70): `resumeStream` never set `requireExistingSession`**, so the headline durable scenario — a restarted process resolving a checkpoint while only the CheckpointStore was wired durably — fabricated a brand-new empty session under the checkpoint's id and pointed a dangling leaf at it. Every other resume-shaped call site already set it. The resulting `resume.session_not_found` now reopens the checkpoint as `env_failed`: the action never ran and the human's decision is still valid, so the approval is retryable rather than burned.
-- **fix (RB-71): one throwing getter collapsed an entire object to a single `<unserializable>` sentinel**, so any two hostile payloads hashed identically — and `boundInputHashOf` exists precisely to prove the args being executed are the args a human saw. The sentinel is now scoped to the offending key; totality is unchanged, resolution costs one key instead of the whole payload.
+- **fix (RB-71): one throwing getter collapsed an entire object to a single `<unserializable>` sentinel**, so any two unserializable payloads hashed identically — and `boundInputHashOf` exists precisely to prove the args being executed are the args a human saw. The sentinel is now scoped to the offending key; totality is unchanged, resolution costs one key instead of the whole payload.
 - **fix (RB-57): `findOrphanToolCalls` resolved tool calls against a global id set**, so a provider that reuses a toolCallId across turns (OpenAI-compatible gateways that number per request) hid a genuine orphan — and an unclosed `tool_use` makes every subsequent request on that session fail at a strict provider, not just one reconcile. Resolution is now positional.
 - **fix (RB-65): `tightenTaskSpec` left `excludeTools`/`deferTools` outside its tighten-only guard** — an override replaced the base list instead of unioning with it, and `[]` cleared every base exclusion outright.
 - **fix (RB-69): the CC task-list store's addressable id set and its enumerable id set were different** — a `007.json` counted toward the allocation high-water yet was unreachable through get/set/delete, and `list()` reported it as the id `"7"`. Checked against CC 2.1.219 itself (`edr`/`Uid`/`ste`): CC constrains a character class, not a numeric shape, and its list returns filenames verbatim, so `007` is a first-class id there. Both sides now use CC's charset; sema keeps the stricter half of the divergence (an off-charset id throws rather than being silently rewritten into a neighbouring file).
@@ -3737,7 +3876,7 @@ design/154 (shellGate classify granularity — server [1558]④): compound read-
 - **feat: `classifyCompoundReadonly(command, allow)`** (exported) — textual scan-split on `;`/`&&`/`||`/`|`, every segment re-vetted by the single shared `parseLeadingCommandName` + bare-name allowlist; all-green ⇒ read-only. Whole-string hard rejects stay for `< > $ ( )` backtick newline CR backslash; a LONE `&` is hard-rejected too (backgrounding outlives the shell and escapes TaskStop lifecycle governance; `|&` rejects via the same rule). Soundness: with escapes/expansion/redirection hard-rejected, quotes can only HIDE connectors (real boundaries ⊆ textual boundaries), so every executed argv[0] is a vetted segment head — a quote-carrying head is parser-rejected, a clean head equals the executed program name. Error direction is strictly false-negative.
 - **feat: `bashReversibilityProbe` upgraded** to the compound classifier (signature unchanged). The `bash_readonly` tool's `effect:"read"` declaration face deliberately KEEPS the strict single-command check (documented fork). The Monitor face intentionally shares the upgraded probe (design/135 G2: same env seam, classification is about the command text; a polled read-only compound stays read-only) — pinned by two wiring tests.
 - **NOT included by design:** no git/`find`/subcommand allowlisting — "zero-flag" vetting operates on pre-expansion text while the shell expands quotes/braces/globs into argv, and the candidate git subcommands are not provably read-only under repo config (`core.fsmonitor`, `diff.external`/textconv, promisor fetches, index refresh). `git status` still asks — recorded false-negative; widening needs a dedicated review.
-- Verification: adversarial review (SOUND across five attack directions) + a black-box harness in the server-[1558] methodology — red-first detector self-check, then 5000 seeded fuzz commands with EVERY green actually executed under real bash in a snapshot sandbox: zero filesystem mutations, zero surviving process groups, zero hangs; curated adversarial family all red or verified-harmless. Ask-rate on a realistic coding-task corpus: 69% → 43%.
+- Verification: adversarial review (SOUND across five probe directions) + a black-box harness in the server-[1558] methodology — red-first detector self-check, then 5000 seeded fuzz commands with EVERY green actually executed under real bash in a snapshot sandbox: zero filesystem mutations, zero surviving process groups, zero hangs; curated adversarial family all red or verified-harmless. Ask-rate on a realistic coding-task corpus: 69% → 43%.
 - Backlog: RB-41 (availability residue — stdin-blocking greens like bare `cat` hang to tool timeout; not a reversibility issue, documented).
 
 ## 1.386.0 (2026-07-23)
@@ -3985,7 +4124,7 @@ CWD case core B-half ([1451]/[1452]/[1461]④): resume workspace-root fidelity +
 - **hardening round 5 (1 high folded): a divergent restore under a PENDING APPROVED ACTION fails closed.** The approved args were authored + human-approved against the checkpointed root — internal path-state rebase cannot rewrite tool args (`rm /app/…` would run against a root the workspace no longer occupies, or burn the approval on a fence refusal). Divergence + pending approval → `resume.env_failed` → the existing reopen leg (resolved→pending) keeps the approval retryable until the adapter honors `priorHandle` or the knob rolls back (pinned end-to-end: fail → reopen → knob-back retry executes exactly once at the right root). The guard keys on the RESUME OUTCOME (`tool_approval` × an `allow` winner — `PrepareResume.executesApprovedAction`), not on the checkpoint kind: a DENY winner executes nothing and resolves under the diverged root without wedging (round 6); non-approval suspends (resource_limit etc.) likewise proceed under the rebase.
 - **hardening round 4 (2 high folded):** the rebase contract is POSIX-ONLY — every remote lane's `mountPath` is a Linux container path; a backslash anywhere marks the value outside the domain and the path passes through UNCHANGED (un-rebased is observable via the divergence observation; wrongly-guessed Windows drive/UNC recomposition would silently corrupt cwd/read-state). And the divergence/misalignment `onError` observations are try/catch-isolated — a throwing observer sink must never turn an already-consumed (resolved) checkpoint into a terminal failure (observe-only means observe-only).
 - **residuals ledgered (RB-35, adversarial round 7, LOW):** two documented tails of the narrow "rotation window × non-honoring adapter × durable suspend" pathology — canonical-vs-logical root prefixes can miss the rebase on symlinked mounts (observable degradation: cd failure / fence refusal / read-state miss, never a silent wrong write), and the fail-closed guard keys on the allow winner before zero-execution branches (walltime exhaustion, tightened rules) are known (a permanent rotation can loop fail→reopen with an explicit remedy in every error). Both wait on a real deployment shape.
-- **test: 13 new pins** — priorHandle rides resumeVM, mount-rotation divergence observation (non-fatal), POSITIVE honor leg (prompt cwd = checkpointed root across a rotation, zero divergence noise), resumeFacts render trio (honest sentences / hostile note / fresh-task-never), resume-leg e2e gating, cwd-alignment guard both ways (honoring env silent; dropping env observed, task proceeds).
+- **test: 13 new pins** — priorHandle rides resumeVM, mount-rotation divergence observation (non-fatal), POSITIVE honor leg (prompt cwd = checkpointed root across a rotation, zero divergence noise), resumeFacts render trio (honest sentences / degraded note / fresh-task-never), resume-leg e2e gating, cwd-alignment guard both ways (honoring env silent; dropping env observed, task proceeds).
 
 ## 1.359.0 (2026-07-20)
 
@@ -4123,7 +4262,7 @@ Redaction tiering ([1295]② — the assertion-killer) + workflow schema teachin
 
 - **fix: same-principal faces stop redacting paths/URLs.** `boundedRedactedSummary` (workflow agent outputs/results/errors flowing back to the launching model+script; run rows behind the scope-gated observe API) now applies `redactSecrets` only — the old blanket `[redacted-path]`/`[redacted-url]` strip killed the cross-agent assertion pattern ("return the path of the file you wrote") while protecting nothing on faces whose isolation is ACCESS CONTROL. CC redacts nothing here; the secret arm stays as a declared superset. Three host-error sites in the Workflow tool (script-path read failures) join the tier — their paths are the HOST'S own error text.
 - **`redactSecrets` rides the centralized `scrubSecrets` scanner** (codex HIGH: a second hand-rolled vocabulary WAS the bypass — SAS `sig=`, JWTs, sk-prefixed path tokens all slipped it): prefixed tokens, JWT, PEM, Bearer, `secretKey=value`, plus two URL arms that matter once URLs survive — userinfo credentials stripped to the LAST `@` of the authority (a first-`@` regex left `p@ssword` half-visible and let an empty-username form escape entirely, codex HIGH), and signed-URL params (`sig`/`signature`/`sas`/`code`) redacted by key.
-- **External faces keep the FULL strip** (codex HIGH): `untrustedEgressForHuman` (human display; URL stripping also disarms exfil-lure links) and the cross-replica `WorkflowCompletionNotifier` payload (its contract promises host paths never escape) still apply `redactHostLeaks`. `subscribeWorkflow` remains in-process/by-runId — a deployment exposing it over HTTP scope-gates first (as `getWorkflowRun` does).
+- **External faces keep the FULL strip** (codex HIGH): `untrustedEgressForHuman` (human display; URL stripping also removes outbound-link bait) and the cross-replica `WorkflowCompletionNotifier` payload (its contract promises host paths never escape) still apply `redactHostLeaks`. `subscribeWorkflow` remains in-process/by-runId — a deployment exposing it over HTTP scope-gates first (as `getWorkflowRun` does).
 - **teaching ([1295]①④)**: the Workflow tool card now states standard-JSON-Schema union support (`{"type":["string","null"]}` / `anyOf` — typebox handles both, verified) and that cross-agent assertions belong in the SCRIPT body (plain JS between stages). Suite green, tsc 0, deepseek live 9/9.
 
 ## 1.346.0 (2026-07-19)
@@ -4204,7 +4343,7 @@ Approval display projection ([1245] — the "Run a dynamic workflow?" confirmati
 - **feat: `approvalPreview` seam** — a tool may declare `approvalPreview(args)` (AgentTool/ToolSpec, mechanism-neutral): a pure display projection of its args for human approval surfaces. The gate mints it through ONE helper — alias-aware tool lookup, throw-swallowed, 16KiB serialized clamp, and **control-character sanitization on every string leaf** (terminal escapes/CR can never reach an approval renderer raw). Delivered on BOTH lanes: `AskRequest.preview` (live onAsk dialogs) and the durable checkpoint's `pendingAction.preview` + **`CheckpointSummary.preview`** (the one-call `listByScope` inbox — no N+1 get).
 - **Trust contract (explicit)**: the preview is UNTRUSTED, ADVISORY display metadata — renderers must contextually escape it and approval surfaces should show the bound raw args alongside (args + boundInputHash remain the sole execution contract; adjudication and resume never read the preview).
 - **run_workflow implements it**: script form → statically parsed meta (name/description/phases[{title,detail}], scriptChars, hasArgs); named/scriptPath forms → identity stubs; malformed meta → bounded parse-error head. Runtime agent prompts are statically unknowable — phase detail is the honest ceiling. Also pinned: run_workflow rides the tool gate with NO exemption (an ask-policy naming it parks/asks like any tool — the [1230] "manual mode ran without confirmation" was a policy name-list gap, not an engine bypass).
-- codex adversarial 2 rounds (hostile-metadata sanitization, alias-invoked preview loss, durable-summary projection all folded). Suite 6181 green, tsc 0, deepseek live 9/9.
+- codex adversarial 2 rounds (malformed-metadata sanitization, alias-invoked preview loss, durable-summary projection all folded). Suite 6181 green, tsc 0, deepseek live 9/9.
 
 ## 1.337.0 (2026-07-19)
 

package/dist/core/auto-mode-prompt.js

@@ -102,5 +102,5 @@ export function renderAutoModeWindow(messages, options) {
 export function renderAutoModeAction(input) {
     const ask = input.askMessage ? `\npermission gate: ${input.askMessage}` : "";
     return (`\n## New action to classify (the agent's most recent action — evaluate THIS)\n\n` +
-        `[tool_call] ${input.req.toolName} ${JSON.stringify(input.req.args ?? {})}${ask}\n`);
+        `[tool_call] ${input.req.toolName} ${excerpt(JSON.stringify(input.req.args ?? {}), 48_000)}${ask}\n`);
 }

package/dist/core/checkpoint-store.d.ts

@@ -1610,6 +1610,12 @@ export declare class CheckpointError extends Error {
      *  Rejected pre-CAS so the checkpoint stays `pending` and a capable worker can still resume it
      *  (design/49 §2/§3, code-ready council round-2). */
      | "checkpoint.unsupported_version"
+    /** backlog #243: `resume`/`resumeStream` was handed `internals.requestedCwd` for a checkpoint that
+     *  carries a remote `workspaceHandle`. The restored workspace's mountPath is authoritative for the
+     *  task root, so the cwd would be ignored at best — and verifying it would touch the env BEFORE the
+     *  VM restore (a lazy-connect adapter then connects a fresh empty sandbox, the design/49 disease).
+     *  Rejected pre-CAS so the checkpoint stays `pending`; re-resume without `requestedCwd`. */
+     | "checkpoint.cwd_conflicts_restore"
     /** design/164: the checkpoint was written by a pre-164 worker AND its ledger carries the RETIRED
      *  cross-slice wall-clock allocation (`resourceLedger.totalWalltimeSec`). The axis no longer exists,
      *  so resuming would run the leg with the operator's time ceiling silently unenforced. Refused
@@ -1733,6 +1739,12 @@ export declare class CheckpointError extends Error {
      *  Rejected pre-CAS so the checkpoint stays `pending` and a capable worker can still resume it
      *  (design/49 §2/§3, code-ready council round-2). */
      | "checkpoint.unsupported_version"
+    /** backlog #243: `resume`/`resumeStream` was handed `internals.requestedCwd` for a checkpoint that
+     *  carries a remote `workspaceHandle`. The restored workspace's mountPath is authoritative for the
+     *  task root, so the cwd would be ignored at best — and verifying it would touch the env BEFORE the
+     *  VM restore (a lazy-connect adapter then connects a fresh empty sandbox, the design/49 disease).
+     *  Rejected pre-CAS so the checkpoint stays `pending`; re-resume without `requestedCwd`. */
+     | "checkpoint.cwd_conflicts_restore"
     /** design/164: the checkpoint was written by a pre-164 worker AND its ledger carries the RETIRED
      *  cross-slice wall-clock allocation (`resourceLedger.totalWalltimeSec`). The axis no longer exists,
      *  so resuming would run the leg with the operator's time ceiling silently unenforced. Refused

package/dist/core/governance-codes.js

@@ -24,6 +24,8 @@ export const NON_GOVERNANCE_MEMORY_CODES = new Set([
     "memory.erasure_selector_mismatch",
     "memory.erasure_census_incomplete",
     "memory.erasure_index_residue",
+    "memory.export_incomplete",
+    "memory.import_rejected",
 ]);
 export function governanceRetryClass(code) {
     if (Object.prototype.hasOwnProperty.call(GOVERNANCE_CODES, code)) {

package/dist/core/hooks.js

@@ -398,7 +398,15 @@ export async function runToolGate(input) {
             denySource = input.shellGated === true ? "shellGate" : "safety";
         }
     }
-    if (input.peerMessage === true && decision.action === "allow" && input.autoMode !== undefined && !input.autoMode.decider.breakerOpen()) {
+    const breakerKnownOpen = (d) => {
+        try {
+            return d.breakerOpen();
+        }
+        catch {
+            return false;
+        }
+    };
+    if (input.peerMessage === true && decision.action === "allow" && input.autoMode !== undefined && !breakerKnownOpen(input.autoMode.decider)) {
         decision = {
             action: "ask",
             message: `tool "${toolName}" sends a message to another agent — routed for classifier review in auto mode`,

package/dist/core/memory-engine/engine.d.ts

@@ -1,4 +1,5 @@
 import { type CommittedBinding, type EraseMemoryEntriesInput, type MemoryErasureAttestation, type TransferEvidence } from "./file-backend.js";
+import { type MemoryExportBundle, type MemoryImportReport } from "./export-bundle.js";
 import { type ChallengeAssignment, type ChallengeEvent, type ControlPlaneRebuildReceipt, type StrictControlPlaneLedger, type ChallengedHistoryRow, type LineagePendingTxn, type LineagePromotion, type MemoryPartitionIncidentSink, type RetrievedAccountRow, type SessionPollutionRecord } from "./layout.js";
 import type { HarvestReport, MemoryAnnouncement, MemoryBackend, MemorySessionHandle, ScanFinding } from "./types.js";
 /**
@@ -438,6 +439,32 @@ export declare class MemoryEngine {
      * reported honestly, never fabricated and never a throw (#196 absence-reports-not-silent-green).
      */
     provenanceOf(entryId: string): Promise<EntryProvenanceAccount>;
+    /**
+     * Export the requested scopes as a GOVERNANCE-COMPLETE bundle (design/178 v2 §4.3 + v2-c):
+     * committed entries (audit-snapshot semantics — shadow-backed rows a disk scan would miss are
+     * in; unavailable content is enumerated, never dressed), the scope-sliced custody chain, the
+     * exported entries' unresolved challenge events and lineage rows, and the named sessions'
+     * pollution markers — all read inside the backend's single-lock export composite, so a bundle is
+     * never a mixed-epoch scene. Refusals are loud and total (`memory.export_incomplete`): there is
+     * no degraded/partial bundle shape — "a truncated package that looks complete" is the one
+     * deliverable this API is forbidden to produce.
+     */
+    exportMemoryScopes(scopes: readonly string[]): Promise<MemoryExportBundle>;
+    /**
+     * Import a bundle (design/178 v2-c §1/§7). Validation is TOTAL and lands nothing on failure
+     * (`memory.import_rejected`): integrity (fixed section-hash key set, per-entry content-rev
+     * recomputation, batch-unique ids/event ids), structure (the governance section is REQUIRED —
+     * there is no entries-only escape hatch), and — when `opts.expectedScopes` is present (a service
+     * endpoint pins its principal's grant set here) — the scope authorization envelope. Everything
+     * after validation runs inside the backend's import composite: the synthetic latch (imported ids
+     * stay withheld from every model-visible read face until governance fully lands), the five
+     * import judgments IN-LOCK, entry landing through the one transaction skeleton, the governance
+     * legs, and the receipt-gated latch release. A crash anywhere converges by re-importing the SAME
+     * bundle; a completed import answers its recorded report idempotently.
+     */
+    importMemoryBundle(bundle: MemoryExportBundle, opts?: {
+        expectedScopes?: readonly string[];
+    }): Promise<MemoryImportReport>;
     /** Host API: challenge every entry the lineage ledger attributes to `sessionId` (post-hoc source
      *  falsification — trustedTools misconfigured, a tool re-classified, late delegation evidence).
      *  Same requestId contract as {@link challengeEntries}. */

package/dist/core/memory-engine/engine.js

@@ -6,6 +6,7 @@ import { inlineUntrusted } from "../untrusted-text.js";
 import { formatMemoryAge } from "../memory-recall.js";
 import { computeEntryRev, parseEntryFile, serializeEntryFile } from "./frontmatter.js";
 import { DEFAULT_MAX_ENTRY_DEPTH, MEMORY_INDEX_FILENAME, canonicalJsonStringify, captureErasureInput, erasureRequestInvalid, erasureSelectHash, scanEntryFiles, } from "./file-backend.js";
+import { assembleMemoryExportBundle, computeMemoryBundleHash, memoryBundleInvalid, } from "./export-bundle.js";
 import { QUARANTINE_DIR, SCAN_FUSE_THRESHOLD, quarantineAndTombstone, readIndexRevs, writeIndexRevs, bumpScanFuse, canonicalize, claimRootScope, clearScanFuse, adoptCanonicalKeyedControlDir, deriveControlPlaneDir, drainMemoryAnnouncements, enqueueMemoryAnnouncement, ensureDirExists, isContainedIn, markSessionPolluted, readSessionPollution, recordRetrievedAccount, writeFileNoFollow, readRetrievedAccount, registerScope, registeredScopes, resolveMemoryEngineRoot, scopeDirFor, appendChallengeEvents, appendLineageAudit, rebuildStrictControlPlaneLedger, isStrictControlPlaneLedgerCorrupt, CHALLENGE_LEDGER_MAX_EVENTS, adjudicateLineagePending, challengedEntryIds, clearLineageForEntries, discardLineagePending, lineageAccountOfEntry, lineageContributionsOfSession, lineageLatchedIds, promoteLineagePending, readChallengeEvents, readChallengedHistory, readLineageRecord, recordChallengedHistory, recordLineageCredential, reconcileLineage, resolveChallengeEvent, stageLineagePending, } from "./layout.js";
 import { scanMemoryFileName, scanMemoryWrite, scanRemediation } from "./scan.js";
 export const MEMORY_INSTRUCTION_TEMPLATE = `# Memory
@@ -521,6 +522,105 @@ export class MemoryEngine {
             return { v: 1, id: entryId, binding, contributors: plane.contributors, ...(plane.exclusion !== undefined ? { exclusion: plane.exclusion } : {}), contentState, ...(ingest !== undefined ? { ingest } : {}), custody };
         }
     }
+    async exportMemoryScopes(scopes) {
+        if (!Array.isArray(scopes) || scopes.length === 0 || scopes.some((s) => typeof s !== "string" || s.length === 0) || new Set(scopes).size !== scopes.length) {
+            const e = new Error("exportMemoryScopes: scopes must be a non-empty array of unique non-empty scope names");
+            e.code = "config.memory_export_request";
+            throw e;
+        }
+        const requested = [...scopes];
+        const face = this.backend.exportSnapshotOf;
+        if (typeof face !== "function") {
+            const e = new Error("exportMemoryScopes: this backend has no export-snapshot capability (exportSnapshotOf) — a governance-complete bundle cannot be assembled from the generic read faces (they cannot fence the five store faces into one epoch), and a governance-less export is the laundering shape this API refuses by design.");
+            e.code = "memory.export_incomplete";
+            throw e;
+        }
+        const snap = await face.call(this.backend, requested);
+        const exportedIds = new Set([...snap.entries.map((e) => e.id), ...snap.contentUnavailable.map((r) => r.id)]);
+        const resolvedGenerations = new Set();
+        for (const e of snap.challenges)
+            if (e.kind === "resolve")
+                resolvedGenerations.add(`${e.entryId}#${e.generation}`);
+        const challenges = [];
+        for (const e of snap.challenges) {
+            if (e.kind !== "challenge" || !exportedIds.has(e.entryId) || resolvedGenerations.has(`${e.entryId}#${e.generation}`))
+                continue;
+            challenges.push({ eventId: e.eventId, entryId: e.entryId, reason: e.reason, at: e.at, ...(e.challengedRev !== undefined ? { challengedRev: e.challengedRev } : {}) });
+        }
+        const lineage = snap.lineage.filter((r) => exportedIds.has(r.entryId));
+        const namedSessions = new Set(lineage.map((r) => r.sessionId));
+        for (const row of snap.custody) {
+            const sessions = row.sessions;
+            if (Array.isArray(sessions))
+                for (const s of sessions)
+                    if (typeof s === "string")
+                        namedSessions.add(s);
+        }
+        const pollutedSessions = snap.pollutedSessions.filter((p) => namedSessions.has(p.sessionId));
+        return assembleMemoryExportBundle({
+            at: this.now(),
+            scopes: requested,
+            storeId: snap.storeId,
+            entries: snap.entries,
+            challenges,
+            pollutedSessions,
+            lineage,
+            custody: snap.custody,
+            residuals: {
+                quarantined: snap.quarantined,
+                unbound: snap.fullStore ? snap.unbound : [],
+                pendingLatch: snap.fullStore ? snap.pendingLatch : [],
+                contentUnavailable: snap.contentUnavailable,
+                quarantineOpaque: snap.quarantineOpaque,
+                unsliceableCustody: snap.unsliceableCustody,
+                unboundCount: snap.unbound.length,
+                pendingLatchCount: snap.pendingLatch.length,
+            },
+        });
+    }
+    async importMemoryBundle(bundle, opts) {
+        let inert;
+        try {
+            inert = JSON.parse(JSON.stringify(bundle));
+        }
+        catch (err) {
+            const e = new Error(`importMemoryBundle: the bundle is not JSON-serializable (${err instanceof Error ? err.message : String(err)})`);
+            e.code = "config.memory_import_request";
+            throw e;
+        }
+        const bad = memoryBundleInvalid(inert, { ...(opts?.expectedScopes !== undefined ? { expectedScopes: opts.expectedScopes } : {}) });
+        if (bad !== undefined) {
+            const e = new Error(`importMemoryBundle: ${bad} — the whole package is refused, nothing landed`);
+            e.code = "memory.import_rejected";
+            throw e;
+        }
+        const b = inert;
+        const face = this.backend.importBundleCommit;
+        if (typeof face !== "function") {
+            const e = new Error("importMemoryBundle: this backend has no bundle-import capability (importBundleCommit) — the import latch/receipt transaction requires it, and an unlatched import would serve entries before their governance lands; nothing was imported.");
+            e.code = "memory.import_rejected";
+            throw e;
+        }
+        const bundleHash = computeMemoryBundleHash(b.integrity.sectionHashes);
+        const custody = [];
+        const seenEv = new Set();
+        for (const row of b.governance.custody) {
+            if (seenEv.has(row.ev))
+                continue;
+            seenEv.add(row.ev);
+            custody.push(row);
+        }
+        const plan = {
+            bundleHash,
+            sourceStoreId: b.storeId,
+            entries: b.entries,
+            challenges: b.governance.challenges,
+            pollutedSessions: b.governance.pollutedSessions,
+            lineage: b.governance.lineage,
+            custody,
+        };
+        return face.call(this.backend, plan);
+    }
     challengeSession(sessionId, reason, requestId) {
         if (typeof requestId !== "string" || requestId === "") {
             const e = new Error("challengeSession: requestId is required (idempotency identity — retries must reuse it; the engine does not mint one)");
@@ -822,7 +922,9 @@ export class MemoryEngine {
             const rec = reconcileLineage(this.controlDir, this.now);
             this.settlePromotions(rec.promoted);
             for (const u of rec.undecidable) {
-                report.warnings.push(`memory lineage transaction ${u.txnId} is unsettled (a crash landed between commit and its durable credential) — ${u.entryIds.length} entr${u.entryIds.length === 1 ? "y is" : "ies are"} latched (memory reads refuse them) until the host adjudicates it (adjudicatePendingLineage)`);
+                report.warnings.push(u.kind === "latch-only"
+                    ? `memory bundle-import latch ${u.txnId} is unsettled (an import did not run to completion) — ${u.entryIds.length} entr${u.entryIds.length === 1 ? "y is" : "ies are"} latched (memory reads refuse them) until the SAME bundle is imported again to completion (importMemoryBundle; adjudication is refused on import latches)`
+                    : `memory lineage transaction ${u.txnId} is unsettled (a crash landed between commit and its durable credential) — ${u.entryIds.length} entr${u.entryIds.length === 1 ? "y is" : "ies are"} latched (memory reads refuse them) until the host adjudicates it (adjudicatePendingLineage)`);
             }
             if (pollutedReason !== undefined && lineageSessionId !== undefined) {
                 const rec2 = this.sessionPollution(lineageSessionId);