@orkestrel/scaffold 0.0.63 → 0.0.65

package/dist/host/AGENTS.md

@@ -68,7 +68,7 @@ configs/  thin target wrappers around root Vite/TypeScript configuration
 - **No nested functions.** Extract function declarations and assignments from bodies. The only exceptions are an anonymous callback passed directly as an argument and an anonymous function returned directly as a result.
 - **Functional core, imperative shell.** Export pure leaves. Keep stateful or defining orchestration as class methods. Classes compose behavior; they do not forward 1:1 to helpers.
 - **No superfluous wrappers.** A wrapper must add a boundary, invariant, composition, translation, lifecycle, or materially narrower contract. Otherwise use or rename the real symbol and update every consumer.
-- **Minimal public API.** Add or substantively expand a capability with its first real consumer; do not speculate. This is a creation gate, never a later visibility gate. Once an intentional reusable capability exists, expose its top-level source exports through the correct environment barrel regardless of which consumers currently use them, so developers receive the same supported mechanisms the package uses. Remove a symbol only when the capability itself should not exist. Prefer one minimal interface and one shared engine, allowing native backend overrides only for genuine faster paths.
+- **Minimal public API.** Add or substantively expand a capability with its first real consumer; do not speculate. This is a creation gate, never a later visibility gate. Once an intentional reusable capability exists, expose its top-level source exports through the correct environment barrel regardless of which consumers use them, so developers receive the same supported mechanisms the package uses. Remove a symbol only when the capability itself must not exist. Prefer one minimal interface and one shared engine, allowing native backend overrides only for genuine faster paths.
 - **No compatibility shims.** This is greenfield. Update every consumer in the same change.
 - **Mechanism, not product policy.** Framework code supplies reusable mechanisms and stops before application decisions.
 - **No polling architecture.** Park idle work on events and abort signals. Yield long work cooperatively.
@@ -101,7 +101,7 @@ configs/  thin target wrappers around root Vite/TypeScript configuration
 7. **Document:** update the guide, examples, and parity contract.
 8. **Verify:** audit discovery, deferrals, and package contents as applicable. Run the required gates and read their actual output before claiming success.
 
-Quality gates before commit, in order. The acceptance gate is the non-mutating variant; run the mutating `lint` and then `format` first only to converge, then prove with the checks. `lint --fix` rewrites code and its output is not formatter-clean, so a `format` that ran before it leaves `format:check` failing on the file `lint` just rewrote:
+Quality gates before commit, in order. The acceptance gate is the non-mutating variant; run the mutating `lint` and then `format` first only to converge, then prove with the checks. `lint --fix` rewrites code and its output is not formatter-clean, so a `format` that ran before it leaves `format:check` failing on the file `lint` rewrote:
 
 ```text
 npm run format:check → npm run lint:check → npm run check → npm run build → npm test

package/dist/host/CLAUDE.md

@@ -53,3 +53,9 @@ follows it. This file adds only what Claude Code does differently, and cannot we
   ChatGPT approval in the browser.
 - `scripts/codex.sh` only reports readiness. It never installs, authenticates, logs out, reads the
   auth cache, or performs a model call.
+- `scripts/deps.sh` runs `npm ci --ignore-scripts` on every session start and resume whose
+  `package-lock.json` digest differs from `node_modules/.orkestrel-lock.sha256`, and a resume
+  follows every turn boundary. After a tracked command changes the lockfile in this checkout,
+  write the lockfile's SHA-256 digest to that marker in the same turn, before any unit or gate runs
+  here. A marker left stale reinstalls `node_modules` under a live suite on the next resume, and the
+  suite reports a missing package that is present a moment later.

package/dist/host/agents/orchestration.md

@@ -138,7 +138,11 @@ Fall back in this order and record the substitution:
   parallelism, independent review, or substantial context justifies it.
 - Dispatch staging, packing, gate-chain invocation, and instrument authorship as units — `builder`
   for a fully specified script, `verifier` for its evidence — each with a brief and an audit like
-  any other unit. Only the commit and the push stay with the Orchestrator.
+  any other unit. The commit and the push stay with the Orchestrator.
+- A release instrument installs, commits, or runs a mutating tree-wide command, which the
+  permission floor bars every role from, so its run is the Orchestrator's tracked command, retained
+  with its log as the authoring unit's acceptance evidence. An upload loop that must run inside a
+  one-time code's life with the user at the keyboard stays with the Orchestrator whole.
 
 ## Roles
 
@@ -265,13 +269,12 @@ edits, formatter and build races, cache phantoms, and validation cross-talk.
 6. After integration, clear shared caches if needed, then have one independent `verifier` run the
    authoritative tree-wide sweep. A writer's self-report never establishes green.
 7. The Orchestrator's own sweep is a writing dispatch and queues behind the units that own those
-   files. A script that fixes one thing across every target is the easiest way to break the
-   serialization rule,
-   because it does not feel like a dispatch — nobody was named, no brief was written, and it
-   finishes in seconds. It still writes into trees a live unit owns, and a unit whose brief it
-   invalidates will repair the same drift the other way and report a state that is already false.
-   Run it before the units, or after them, or send the decision to every unit in flight per the
-   mid-campaign rule under **Dispatch anatomy**. Never beside them.
+   files. A script that fixes one thing across every target is the most direct way to break the
+   serialization rule, because it does not feel like a dispatch — nobody was named, no brief was
+   written, and it finishes in seconds. It still writes into trees a live unit owns, and a unit
+   whose brief it invalidates will repair the same drift the other way and report a state that is
+   already false. Run it before the units, or after them, or send the decision to every unit in
+   flight per the mid-campaign rule under **Dispatch anatomy**. Never beside them.
 8. A fleet pass that records a per-target status commits only the targets it recorded green. Reading
    "is the tree dirty" instead of "did this target pass" pushes a red target the moment one exists,
    and a flake makes that look like it worked. Refuse the failed row, name it, and re-run it alone
@@ -301,7 +304,7 @@ each CLI first (`codex --version`; `agent --version`, falling back to `agent.cmd
 run the bench's authentication-state check where it exposes one. Neither answer is liveness. A
 version string proves the binary is installed, and an authentication-state check reads stored
 credentials, so both pass while the account is out of quota, while the routed model is unavailable to
-it, while the server has already revoked the credential the check just read, and inside a sandbox
+it, while the server has already revoked the credential the check read, and inside a sandbox
 with the network denied. Record a bench live only on a bounded round-tripped model call that came
 back, and record what came back beside the routing decision. Probes are read-only, and the role file
 owns each bench's exact probe.
@@ -362,7 +365,7 @@ longer holds.
      plan did not consider; **unchanged**.
    - Redraw the dependency order. Strike a unit whose subject a later unit deletes. State the new
      prerequisite of a unit that acquired one.
-   - Walk the remaining units once and ask of each whether what just landed still supports it. A
+   - Walk the remaining units once and ask of each whether what landed still supports it. A
      decision taken inside a unit can remove a later unit's foundation, and the unit that took it
      cannot see that.
    - Re-baseline when a probe overturns a decision the plan rests on, not only at a phase boundary.
@@ -427,6 +430,8 @@ The harness bridge names the concrete mechanism for each of these.
   so a report living only in the Orchestrator's context stops the next lane on arrival.
 - Capture the unit's returned report to a file beside its brief under the same unit name, so a
   unit's instruction and its outcome are one pair on disk.
+- Name, inside a report that rests on a bench lane, that lane's journal path and session id, so
+  the provenance survives the journal's sweep.
 - Amend a brief on re-run rather than restating it. A mid-campaign correction produces a successor
   file recording what changed and why, and the original stays. A fix round's brief names the
   findings it carries and where each came from.
@@ -604,7 +609,8 @@ command that outlives the turn that started it. Every law here binds all of them
   byte offset rather than loading it, so an edit that shifts line numbers moves the text under that
   offset and the shell resumes mid-construct. The run dies on a syntax error in a line the script
   does not contain, which reads as a defect in the work rather than as the edit that caused it. Copy
-  the file, edit the copy, and launch the copy for the next run.
+  the file, edit the copy, and launch the copy for the next run. A successor's header names its own
+  file, its own log, and what changed from the file it supersedes.
 - On a Windows host this binds every program-carrying command, not only long ones. Heredocs,
   `node -e`, `node -p`, `&&` chaining, and any argument carrying `${...}` trip the Git Bash
   approval classifier and turn an unattended run into a manual approval prompt. Write the program
@@ -848,9 +854,9 @@ either publishes packages nobody needed to publish or leaves a consumer pinned t
 
 Every package is `0.0.x`, where a caret pins one exact release. A dependent therefore sees a new
 version only after it re-pins and republishes, so the fleet publishes in topological layer order
-derived from runtime `dependencies` alone. Layers exist for a reason a flat pass cannot fix:
-ranges that disagree install duplicate copies of the same package, and the compiler reads them as
-distinct types.
+derived from the runtime `dependencies` and `peerDependencies` edges, never from a development
+edge. Layers exist for a reason a flat pass cannot fix: ranges that disagree install duplicate
+copies of the same package, and the compiler reads them as distinct types.
 
 Read the order from the catalog table in `.claude/agents/orkestrel.md`, which `scaffold catalog`
 regenerates from the registry. Its `Layer` column is the publish round. Regenerate it before
@@ -861,7 +867,9 @@ The tooling packages sit outside that order because nothing depends on them at r
 is a development dependency of every package, including packages it depends on itself, so a runtime
 layering would report a cycle that does not exist. Each package builds against the already-published
 `scaffold`, never against an unpublished one, and a `scaffold` release therefore publishes on its own
-and propagates as files rather than as a cascade.
+and propagates as files rather than as a cascade. A package the fleet consumes as a development
+dependency takes the same shape when its consumers' gates read its unpublished tip;
+`.agents/skills/orkestrel-publish/references/wave.md` § Rule on the bump states that trigger.
 
 `scaffold` also carries a second published surface beside `dist/src`: `package.json` ships
 `dist/host`, the vendored file set every target receives through `repair`.