@event4u/agent-config 9.24.0 → 9.26.0

package/.claude-plugin/marketplace.json

@@ -6,7 +6,7 @@
   },
   "metadata": {
     "description": "Bootstrap shim for the event4u agent-config suite. This plugin ships ONLY the dispatcher hooks plus one install-pointer skill — ALL content (skills, rules, commands) is installed via `npx -y @event4u/agent-config init`, which writes the file projection AND registers the same hooks in ~/.claude/settings.json. DEPRECATED as a content channel; kept so the marketplace listing stays discoverable and existing installs keep working hooks.",
-    "version": "9.24.0",
+    "version": "9.26.0",
     "keywords": [
       "agent-config",
       "skills",

package/CHANGELOG.md

@@ -303,6 +303,142 @@ Entry-shape contract: [`docs/contracts/CHANGELOG-conventions.md`](docs/contracts
 > that forces a new era split (`# Era: 9.24.x`, etc.) — see
 > [`docs/contracts/CHANGELOG-conventions.md § Era splits`](docs/contracts/CHANGELOG-conventions.md).
 
+## [9.26.0](https://github.com/event4u-app/agent-config/compare/9.25.0...9.26.0) (2026-08-07)
+
+### Release highlights
+
+<!-- Curated head: fill before merge, keep it under 10 lines, and leave `_none_` where it is genuinely the answer. The generated log below is unchanged. -->
+- **Behaviour changes:** _auto-derived, rewrite before merge:_ rule/schema diffs, breaking commits or removed public surface in 17c3716, 27e6649, f352c7a, 8fdcc7b, 581dfa0, a396442 +1 more.
+- **Default changes + migration:** _auto-derived, rewrite before merge:_ commits naming a default, migration or migrate in 3bae006.
+- **Security and correctness:** _none_
+- **Honest nulls:** _auto-derived, rewrite before merge:_ commits carrying an honest-null marker in cefcd37.
+- **Known limitations:** _none_
+
+### Features
+
+* **memory:** gate the session-end learning hook on the recorded consent ([79a6aa8](https://github.com/event4u-app/agent-config/commit/79a6aa892f2d772254ef91c0b19cb93007da2d70))
+* **settings:** resolve the nickname prefill through the documented chain ([e98a7cb](https://github.com/event4u-app/agent-config/commit/e98a7cba5cc76ccd62ab7b84be4b2a2b749d96e5))
+* **bench:** price the cost axis per token bucket in the v2 report ([b234b3e](https://github.com/event4u-app/agent-config/commit/b234b3e7076d401f9201610545528097e84490f2))
+* **bench:** preserve per-trial workspaces and add a real no-network selftest ([12df43b](https://github.com/event4u-app/agent-config/commit/12df43bc6e1ac2be87e62128accc00d1296db991))
+* **bench:** add the package-ladder and bare-principle arms ([c3f1c31](https://github.com/event4u-app/agent-config/commit/c3f1c316c270aa166ed3164165a02c083345319c))
+* **gates:** assert every authored skill REACHES the Claude tree, not just that it parses ([c42b1ad](https://github.com/event4u-app/agent-config/commit/c42b1adb2afb237d45182a59af705ef64eb4fe66))
+* **settings:** make the provenance stamp decide something, and condense the rule ([27e6649](https://github.com/event4u-app/agent-config/commit/27e664999a8894e1b146a556fa114daee3511676))
+* **settings:** normalise the just-in-time settings ask and compute its budget ([f352c7a](https://github.com/event4u-app/agent-config/commit/f352c7aea72beab330f1abd625041f7bcb46df3a))
+* **language-mirror:** fall back to the system locale on a terse first prompt ([784ee64](https://github.com/event4u-app/agent-config/commit/784ee64307086a28e9657b0a0ae5ae2fac94309a))
+* **gates:** fail the pre-push chain when the running hook bundle predates its sources ([a829a7c](https://github.com/event4u-app/agent-config/commit/a829a7cb50fac0fe02fc4af80c7083a890312913))
+* **review:** one findings schema both tracks validate against ([581dfa0](https://github.com/event4u-app/agent-config/commit/581dfa079ee333edcba920b464a19c11542db4c8))
+* **review:** make prompt provenance and author-vs-reviewer checkable ([cefcd37](https://github.com/event4u-app/agent-config/commit/cefcd37ae8a4b0ee5e42ee0514e9b09399f98870))
+* **gate:** enforce the corpus refresh cadence and open the CSVs nobody opened ([c946769](https://github.com/event4u-app/agent-config/commit/c9467697823b5a0b82d8f3e4b4d34c8bd8fe813f))
+* **rules:** close the write side of the UI loop ([a396442](https://github.com/event4u-app/agent-config/commit/a396442da455b6e02afddacc87c462f80b198499))
+* **hooks:** per-concern tools: filter, applied by the dispatcher ([28a8f96](https://github.com/event4u-app/agent-config/commit/28a8f96ee413083e3f451f1bf49b5588483ae9aa))
+* **gate:** ratchet gate_self_test adoption as a shrink-only non-adopter count ([99e3656](https://github.com/event4u-app/agent-config/commit/99e36560c208904510bc0525122ad2549d697bea))
+* **agents:** create agents/proposals — the directory seven artefacts pointed at ([64717a9](https://github.com/event4u-app/agent-config/commit/64717a96680cc70d3de7c4dd9a853d92952fbb53))
+* **design:** route a capability URL and a design-system dir to design-fidelity ([85b1562](https://github.com/event4u-app/agent-config/commit/85b1562c193f9eaf6ed63f8692f79e0b9d2eda3f))
+* **commands:** add /roadmap:next — screen, select, then ship a roadmap ([459211c](https://github.com/event4u-app/agent-config/commit/459211cadcc7682f26c8b081ac6c87c2a4712e46))
+
+### Bug Fixes
+
+* **release:** stop masking every push failure as a moved remote ref ([47faa19](https://github.com/event4u-app/agent-config/commit/47faa19e080bbe27c3cd8dfeb0bee67a10acfbf8))
+* **settings:** retract an overclaim and take the fork off the request path ([2139b17](https://github.com/event4u-app/agent-config/commit/2139b174c8e4afad3f7294bfc196ad0e79801cec))
+* **settings:** close the R2 review findings on the prefill and the pins ([fe35e63](https://github.com/event4u-app/agent-config/commit/fe35e63702661f34e55345c5c4fa7e2cd6b34fa4))
+* **tests:** narrow the ARMS lookup for noUncheckedIndexedAccess ([3b3e1b1](https://github.com/event4u-app/agent-config/commit/3b3e1b14ee8bb588004410d901835d35649b2544))
+* **roadmaps:** make solution-minimalism report its own spend gate ([3d9ceef](https://github.com/event4u-app/agent-config/commit/3d9ceef6153ed6df35292bf74d9a48b1e83dd982))
+* **cli:** both routing commands were dead through the shipped binary ([c2b38e2](https://github.com/event4u-app/agent-config/commit/c2b38e2263e9b27401ea38bb2782b0d7532c4527))
+* **gates:** emit the freshness verdict via stdout, not console.log ([fa06eea](https://github.com/event4u-app/agent-config/commit/fa06eea18620e671f3c09ac877cb094c6022c0aa))
+* **hooks:** design-slop read the tool fields where the dispatcher does not put them ([3ca190f](https://github.com/event4u-app/agent-config/commit/3ca190f0b2364c785f791dc106086209d86f1983))
+* **ci:** close the five surfaces this change reds, one of them a real gate gap ([c7c25ff](https://github.com/event4u-app/agent-config/commit/c7c25ffc1eac8f8b73776c5655f35af42f9eb9f7))
+* **conformance:** round 2 — the shipped gates did not fire, and four reasons were wrong ([8fdcc7b](https://github.com/event4u-app/agent-config/commit/8fdcc7b9d2de7c3239af7e809e86933e41cf44e2))
+* **gate:** close the word-form count hole in lint_abstraction_thresholds ([d5c294c](https://github.com/event4u-app/agent-config/commit/d5c294c4cf547696bf452a385ea0935b191283cd))
+* **video:** refuse --crossfade instead of silently delivering a hard cut ([6eea400](https://github.com/event4u-app/agent-config/commit/6eea4000146d1de2c8de703e046aa8072e9e4d96))
+* **ci:** close the two count-and-fixture surfaces a new sub-command reds ([f8d9566](https://github.com/event4u-app/agent-config/commit/f8d9566ca568c45c18cb2ec62d72492eaecf7570))
+
+### Reverts
+
+* **gates:** drop the skill-completeness check — its premise is false and it blocks the release ([5473a92](https://github.com/event4u-app/agent-config/commit/5473a925e13989fc6ba382145ab20c485dae2995))
+
+### Documentation
+
+* **roadmap:** reopen step 4.2 — the prefill shipped, the ask did not ([5fb6c05](https://github.com/event4u-app/agent-config/commit/5fb6c053a587440749af0b6d29b37c00fc1c03d1))
+* **roadmap:** close Phases 4 and 5 from evidence, re-encode the blocked steps ([2c70288](https://github.com/event4u-app/agent-config/commit/2c70288e372254be0cd0a23acfdd76f3b1318cb1))
+* **bench:** add the ab-v2 reproduce path and the Phase-3 pre-registration ([bb0ad48](https://github.com/event4u-app/agent-config/commit/bb0ad480deddf4ee3739d80d795a6c8546de9ffc))
+* cancel P1.2 on measurement, close the P2.1 null-scope check ([81fce51](https://github.com/event4u-app/agent-config/commit/81fce5189b88b9d64051cbf9d58e4322a783ca57))
+* **roadmap:** close P1.3 and P5.3 with their measured departures ([1c101db](https://github.com/event4u-app/agent-config/commit/1c101db9a045cc0b298927ef3d8d2622e8ec6625))
+* **roadmap:** close the four verified-fix steps of the inbox harvest ([b0cd2f4](https://github.com/event4u-app/agent-config/commit/b0cd2f4773b0f9a97135ecc5ff2432e0ae2513ab))
+
+### Refactoring
+
+* **settings:** point the four bespoke ask sites at the shared protocol ([17c3716](https://github.com/event4u-app/agent-config/commit/17c371647eddc5945dbbcc03e1cb2e698c63cf55))
+
+### Tests
+
+* **settings:** pin the ask-shaped settings migration ([3bae006](https://github.com/event4u-app/agent-config/commit/3bae00676daa7ae0e8c195fd2b6e00a2a25c60bb))
+
+### Chores
+
+* **ci:** retrigger pull_request event delivery ([a20088f](https://github.com/event4u-app/agent-config/commit/a20088f328d533827fac1de60273a3f206aceef1))
+* **roadmaps:** record the cost endpoint against the Endpoints step ([a3585ad](https://github.com/event4u-app/agent-config/commit/a3585ad7152b64fbe250ea294d242ab5a86a693b))
+* **roadmaps:** close the thresholds step, record the reproducibility residue ([39a93d3](https://github.com/event4u-app/agent-config/commit/39a93d33a3e69b132f643fe97380e2a7804d1acb))
+* **reports:** refresh the originality report round 2 left stale ([a65f929](https://github.com/event4u-app/agent-config/commit/a65f92939bae5b170a3cf58464579ccf774c5a9e))
+* regenerate router.json after the trunk merge ([0af3e84](https://github.com/event4u-app/agent-config/commit/0af3e8425106ccc395be79f66faacc3f8b45a5b1))
+* regenerate router.json and the routed-rule count for the new rule ([f262b83](https://github.com/event4u-app/agent-config/commit/f262b83fdba6b71905b03333fd93f657ae0905a3))
+* regenerate the tool trees and indices for the new rule ([2788940](https://github.com/event4u-app/agent-config/commit/2788940cc55326e125c2000ad3cf0e90fe2b5c51))
+* **router:** regenerate router.json for the two new design-fidelity triggers ([20aee52](https://github.com/event4u-app/agent-config/commit/20aee52cb1d440d77ec4b9e712164addfd0a83de))
+
+### Other
+
+* **settings:** record the R2 findings, and announce the Iron Law 3 recurrence ([63e2f09](https://github.com/event4u-app/agent-config/commit/63e2f094c57c77ec18ecf4cf449e7682597be29f))
+
+Tests: 11643 (+200 since 9.25.0)
+
+## [9.25.0](https://github.com/event4u-app/agent-config/compare/9.24.0...9.25.0) (2026-08-06)
+
+### Release highlights
+
+<!-- Curated head: fill before merge, keep it under 10 lines, and leave `_none_` where it is genuinely the answer. The generated log below is unchanged. -->
+- **Behaviour changes:** _auto-derived, rewrite before merge:_ rule/schema diffs, breaking commits or removed public surface in 5d653e2, 00018c4, 20bb411, cb0917d, 7641291.
+- **Default changes + migration:** _auto-derived, rewrite before merge:_ commits naming a default, migration or migrate in bd45f62, 11a699c.
+- **Security and correctness:** _none_
+- **Honest nulls:** _auto-derived, rewrite before merge:_ commits carrying an honest-null marker in 5d653e2, 20bb411.
+- **Known limitations:** _none_
+
+### Features
+
+* **commands:** add /analyze:conformance, and clear the six red CI checks ([5d653e2](https://github.com/event4u-app/agent-config/commit/5d653e2ecfa1b697c7c44a94087d0248debea3d9))
+* **cli:** add conformance:behavior — replay transcripts through the shipped gates ([7c7a9dc](https://github.com/event4u-app/agent-config/commit/7c7a9dc10d83245b7507cbcef348e52901968c45))
+* **hooks:** gate the language pin, git authorization and evaluator independence ([582a9f3](https://github.com/event4u-app/agent-config/commit/582a9f37b97d3cb1f6474963fea576349f6b0b5d))
+* **authoring:** a decision-impact class, and four conventions Phase 2 owed ([7641291](https://github.com/event4u-app/agent-config/commit/7641291866dac3c24deaf4cea1827a6c24d63edb))
+* **ledger:** record what the rule-body migration kept, folded, and lost ([11a699c](https://github.com/event4u-app/agent-config/commit/11a699cda54c2eca55899af25435aee80a6b7f75))
+* **skill-writing:** two conventions that travel with the artifact ([9120d0f](https://github.com/event4u-app/agent-config/commit/9120d0ffd1d8ea7140f9cc393e34c595dba29085))
+* **interactions:** close the register over the set it claims to cover ([654837b](https://github.com/event4u-app/agent-config/commit/654837b1b8c3e33a43539d256bcd18592ade6621))
+* **descriptions:** reject a description that argues against its siblings ([9889529](https://github.com/event4u-app/agent-config/commit/9889529bcb34e604fd2bedde3ab2abf98e3f6a40))
+* **gates:** lint authored examples, and cap the on-demand depth layer ([c2c2d91](https://github.com/event4u-app/agent-config/commit/c2c2d917cede10eab64144ac9302857f6edd80b5))
+
+### Bug Fixes
+
+* **rules:** clear a lint regression and register the new rule in the router ([00018c4](https://github.com/event4u-app/agent-config/commit/00018c4315269467b856713ec056f2c61982ca33))
+* **roadmap:** drop the duplicate un-archived copy ([532200a](https://github.com/event4u-app/agent-config/commit/532200ac94966a83df0a97e0c48e02e240d72c3f))
+* **hooks:** stop counting vacuous verification as evidence ([7696e54](https://github.com/event4u-app/agent-config/commit/7696e5453b770d200e2925204cba4e08f323ca7b))
+* **ci:** bump the gate-script population and regenerate the artefact index ([21a16f1](https://github.com/event4u-app/agent-config/commit/21a16f15a04c1c6b515832ddf56a951fce6b16d1))
+* **schema:** insert decision_impact surgically instead of reserialising the file ([cb0917d](https://github.com/event4u-app/agent-config/commit/cb0917daca7080b2e5446473b6e420b854554d5e))
+
+### Documentation
+
+* **analyze-conformance:** adopt the checked-out-branch default from its sibling ([bd45f62](https://github.com/event4u-app/agent-config/commit/bd45f62e52edb86fd4e22cc8828a819a60788ae7))
+* **analyze-inbox:** run in the checked-out branch; worktree only on request ([aaf29b4](https://github.com/event4u-app/agent-config/commit/aaf29b466e4df37236e32fd588077c83f51f5483))
+* **roadmap:** record the conformance audit and close its roadmap ([1616c23](https://github.com/event4u-app/agent-config/commit/1616c236d4e47b0bc77af6521a2db23419760f75))
+* **rules:** add evaluator-independence, and downgrade two enforcement claims ([20bb411](https://github.com/event4u-app/agent-config/commit/20bb4116d7d0a2dec198d4e08ade4371ff802b8a))
+* **roadmap:** close and archive the authoring-discipline roadmap ([977ebd9](https://github.com/event4u-app/agent-config/commit/977ebd9a3771eaeb42b631cd9660bf4b16d0d1ae))
+* **roadmap:** record the second execution pass and its three measured departures ([ec1acfb](https://github.com/event4u-app/agent-config/commit/ec1acfbfc5718a2197bdc9ca5cdcc091ba3a410c))
+* **roadmap:** close Phase 4 steps 1-3 and 5, with what changed from the plan ([87d4bd2](https://github.com/event4u-app/agent-config/commit/87d4bd24e5f2b7f1e7429ba016e7d473aec00fa7))
+
+### Chores
+
+* **docs:** regenerate the artefact index for /analyze:conformance ([d1f8f25](https://github.com/event4u-app/agent-config/commit/d1f8f25105b8df849160b2b62e78a84f555d8222))
+* **docs:** regenerate the artefact index for the new rule ([910555a](https://github.com/event4u-app/agent-config/commit/910555a46f5345352e8dd6a950a370ac53019fa1))
+* **sync:** regenerate the projections for the two authoring skills ([44f037e](https://github.com/event4u-app/agent-config/commit/44f037efc1d92d07f6efed30ecb4cb612b2bb55a))
+
+Tests: 11443 (+134 since 9.24.0)
+
 ## [9.24.0](https://github.com/event4u-app/agent-config/compare/9.23.0...9.24.0) (2026-08-06)
 
 ### Release highlights

package/README.md

@@ -4,7 +4,7 @@
 
 [![Smoke](https://github.com/event4u-app/agent-config/actions/workflows/smoke.yml/badge.svg)](https://github.com/event4u-app/agent-config/actions/workflows/smoke.yml) [![Public install smoke (3 OS × 2 Node)](https://github.com/event4u-app/agent-config/actions/workflows/smoke-public-install.yml/badge.svg)](https://github.com/event4u-app/agent-config/actions/workflows/smoke-public-install.yml) [![npm](https://img.shields.io/npm/v/@event4u/agent-config?style=flat-square&label=npm&color=orange)](https://www.npmjs.com/package/@event4u/agent-config) [![agent-config MCP server](https://glama.ai/mcp/servers/event4u-app/agent-config/badges/score.svg)](https://glama.ai/mcp/servers/event4u-app/agent-config) [![MCP Toplist](https://mcptoplist.com/badge/glama%2Fevent4u-app%2Fagent-config.svg)](https://mcptoplist.com/server/glama%2Fevent4u-app%2Fagent-config)
 
-[![Skills](https://img.shields.io/badge/Skills-288-orange?style=flat-square)](dist/agent-src/skills/) [![Rules](https://img.shields.io/badge/Rules-111-orange?style=flat-square)](dist/agent-src/rules/) [![Commands](https://img.shields.io/badge/Commands-194-orange?style=flat-square)](dist/agent-src/commands/) [![Guidelines](https://img.shields.io/badge/Guidelines-104-orange?style=flat-square)](docs/guidelines/) [![Personas](https://img.shields.io/badge/Personas-29-orange?style=flat-square)](dist/agent-src/personas/) [![Advisors](https://img.shields.io/badge/Advisors-5-orange?style=flat-square)](dist/agent-src/personas/advisors/)
+[![Skills](https://img.shields.io/badge/Skills-288-orange?style=flat-square)](dist/agent-src/skills/) [![Rules](https://img.shields.io/badge/Rules-114-orange?style=flat-square)](dist/agent-src/rules/) [![Commands](https://img.shields.io/badge/Commands-196-orange?style=flat-square)](dist/agent-src/commands/) [![Guidelines](https://img.shields.io/badge/Guidelines-104-orange?style=flat-square)](docs/guidelines/) [![Personas](https://img.shields.io/badge/Personas-29-orange?style=flat-square)](dist/agent-src/personas/) [![Advisors](https://img.shields.io/badge/Advisors-5-orange?style=flat-square)](dist/agent-src/personas/advisors/)
 
 **Try it in 30 seconds** — drop one read-only subagent into any repo and watch it gate "done": `@production-validator check this branch is actually done`. No wizard, no lock-in, nothing else installed — the [30-second wedge ↓](#quickstart) is the whole first step. Start at the proof, not the catalog: **[event4u-app.github.io/agent-config/proof/](https://event4u-app.github.io/agent-config/proof/)**.
 

package/dist/agent-src/commands/analyze/conformance.md

@@ -0,0 +1,182 @@
+---
+model_tier: high
+name: analyze-conformance
+pack: analysis-workbench
+tier: 2
+visibility: internal
+sub: conformance
+cluster: analyze
+skills: [roadmap-writing, decision-review, ai-council, subagent-orchestration]
+description: Audit recent local sessions for rule violations — deterministic scan plus subagent passes over the transcripts, root-cause each class, and emit a roadmap that mechanises what is mechanisable.
+argument-hint: "[--limit N] [--scan-only] [--no-council] [--no-roadmap] [--worktree]"
+suggestion:
+  eligible: false
+  rationale: "Cluster sub-command — reached via its cluster head's routing or its explicit /analyze:conformance name; not independently suggested (surface-consolidation)."
+workspaces:
+  - agent-config-maintainer
+packs:
+  - meta
+---
+
+# analyze-conformance
+
+Answer one question about this agent, from evidence rather than impression:
+**where did it not work the way its own rules require, and why?**
+
+The sibling [`/analyze:inbox`](../inbox/command.md) turns somebody else's opinion
+into a roadmap. This turns **the agent's own recorded behaviour** into one. The
+input is not a document — it is the local transcript store, which is the only
+place the agent's actual conduct is written down.
+
+## Why this is not just a grep
+
+The first version of this audit measured 303 language violations. The real
+number was 626. The detector had inherited the very defect it was measuring:
+it classified "the most recent user-role entry", and slash-command and skill
+bodies arrive in the `user` role. It scored the worst session — 136 of 136
+assistant turns in the wrong language — at **3**.
+
+That is the shape of this whole job. A conformance audit is measuring a
+measurement failure, so every instrument has to be checked against a case whose
+answer is already known before its number is believed.
+
+## Where the work happens — the checked-out branch, unless asked otherwise
+
+```
+RUN IN THE CURRENT BRANCH. NEVER CREATE A BRANCH, A WORKTREE, OR A PR
+ON THIS COMMAND'S OWN AUTHORITY. A WORKTREE ONLY ON AN EXPLICIT ASK.
+```
+
+Same default and same reasoning as [`/analyze:inbox`](../inbox/command.md): the
+roadmap and the findings land in the branch already checked out, as ordinary
+uncommitted changes for the operator to review. This command authorizes
+**analysis and authoring** — not a git shape
+([`scope-control`](../../../../rules/scope-control.md) § Git operations).
+
+An isolated worktree runs only when the invocation asks for one — `--worktree`,
+or the operator saying so in the prompt. Then follow
+[`using-git-worktrees`](../../../../skills/using-git-worktrees/SKILL.md),
+including its seeding allow/deny list.
+
+One difference from its sibling worth knowing: a **full** run of this command
+(subagent fan-out plus council) is long enough that a worktree can genuinely
+earn its price by keeping a multi-file diff off the current branch. It still
+does not take one by itself — that is the operator's call, and `--worktree` is
+how they make it.
+
+## Procedure
+
+### 1 — Scan (deterministic, always)
+
+```bash
+agent-config conformance:behavior --limit <N> --output conformance-report.json
+```
+
+Replays the local transcript store through the four **mechanised** checks:
+language pin, git authorization, vacuous evidence, evidence steering. Every
+classifier is imported from the gate it measures, so the scan and the gate
+cannot drift apart.
+
+This is the floor, not the audit. It sees only what already has a gate — which
+is exactly why it is trustworthy and exactly why it is not sufficient.
+
+`--scan-only` stops here.
+
+### 2 — Read the transcripts (subagent fan-out)
+
+The scan cannot see a rule that has no gate: an ask handed over as a trailing
+parenthetical, a completion claimed on the wrong evidence, a root cause named
+and then not fixed. Those need reading.
+
+- Build per-session digests (user turns + assistant prose, tool calls stripped)
+  so a session fits in a subagent's context.
+- Group them into balanced batches by size, and dispatch **one subagent per
+  batch, in parallel**, each returning findings in a fixed shape:
+  rule · session · timestamp · **verbatim quote** · what happened · why
+  (hypothesis) · severity.
+- A finding without a quote is discarded. The quote is what makes the finding
+  auditable by someone who was not in the session.
+
+Highest-signal evidence, in order: a user turn correcting the agent; a user turn
+expressing frustration; the agent's own retro sentence admitting a mistake. Any
+of those three outranks a rule read in the abstract, because it is ground truth
+about what actually cost the user something.
+
+### 3 — Root-cause each class, then pick a mechanism
+
+Group findings into classes and, for each, ask the only question that matters:
+
+> The rule was in context, at full strength, when this happened. So what would
+> a prose change buy?
+
+Usually nothing. Choose exactly one per class:
+
+| Choice | When |
+|---|---|
+| **Deterministic gate** | The failure is a state transition observable at a hook event (a git op, a claim of verification, a dispatch). Name the event and whether it warns or blocks. |
+| **Context injection** | The context is genuinely missing a *fact*, not attention — the language pin is the worked example. |
+| **Rule-text change** | The rule is wrong or its trigger is mis-stated, not merely unheeded. |
+| **Nothing** | Accept the rate and **delete the claim that it is enforced.** |
+
+Reach for a mechanism only where the failure mode was measured. A gate for a
+class nobody was harmed by is the mechanism-without-a-failure-mode this repo
+forbids.
+
+### 4 — Council (unless `--no-council`)
+
+Route the class list and the proposed mechanisms to the AI council. Ask it
+directly for the strongest objection to your own plan, and ask which parts are
+theatre. Record where you **depart** from its verdict and why — a council
+recommendation adopted without a stated reason is not a review, it is cover.
+
+### 5 — Roadmap (unless `--no-roadmap`)
+
+Per [`roadmap-writing`](../../../../skills/roadmap-writing/SKILL.md). The
+roadmap must carry:
+
+- the measured numbers, with the method that produced them;
+- any number that was **corrected during the audit**, published beside the
+  correction rather than silently replaced;
+- one phase per mechanised class;
+- a final phase of **honest downgrades** — every enforcement claim the audit
+  proved false;
+- a table of classes deliberately **not** fixed, each with its reason.
+
+### 6 — Validate the instruments before trusting them
+
+Before the roadmap is believed, run every new detector and gate against:
+
+- **the worst known case** — if it does not fire there, the instrument is wrong,
+  not the corpus;
+- **this session's own behaviour** — the audit that produced this command found
+  its own evidence-steering gate flagging six of its own analysis subagents.
+
+## Flags
+
+| Flag | Effect |
+|---|---|
+| `--limit N` | How many recent sessions to read (default 30). |
+| `--scan-only` | Step 1 only — the deterministic report, no subagents, no spend. |
+| `--no-council` | Skip step 4. Record that it was skipped. |
+| `--no-roadmap` | Findings only; no artefact authored. |
+| `--worktree` | Opt into an isolated worktree. Off by default — see above. |
+
+## Do NOT
+
+- Trust a number this audit computed without validating the instrument (§ 6).
+- Ship a stronger adjective as a fix for a rule that already failed at full
+  strength.
+- Let the scan cover a class that has no gate — a report over un-mechanised
+  rules reads as enforcement and is not.
+- Fold execution into this command. It audits and it plans; running the roadmap
+  is [`/roadmap:process-full`](../../../meta/roadmap/process-full/command.md) on
+  a later turn, per `scope-control` § Authoring vs. implementation.
+
+## See also
+
+- [`conformance:behavior`](../../../../scripts/conformance_scan.ts) — the
+  deterministic half, and the only part that runs without a model.
+- [`/analyze:inbox`](../inbox/command.md) — same output shape, external input.
+- [`evaluator-independence`](../../../../rules/evaluator-independence.md) —
+  binds step 2: this command dispatches evaluators, so its own prompts are
+  subject to it.

package/dist/agent-src/commands/analyze/inbox.md

@@ -8,7 +8,7 @@ sub: inbox
 cluster: analyze
 skills: [learning-to-rule-or-skill, roadmap-writing, decision-review]
 description: Analyze a dropped inbox artifact (review, prompt, spec, transcript) against the current tree, verify every claim it makes, map what survives onto this suite's artefact types, emit a roadmap each.
-argument-hint: "[<file> | <dir>] [--triage-only] [--no-roadmap] [--keep-inbox]"
+argument-hint: "[<file> | <dir>] [--triage-only] [--no-roadmap] [--keep-inbox] [--worktree]"
 suggestion:
   eligible: false
   rationale: "Cluster sub-command — reached via its cluster head's routing or its explicit /analyze:inbox name; not independently suggested (surface-consolidation)."
@@ -40,7 +40,35 @@ still true.
 
 Flags: `--triage-only` stops after Phase 2 (no deep analysis, no roadmap) ·
 `--no-roadmap` analyses but emits findings instead of roadmaps ·
-`--keep-inbox` skips the `tmp.old/` move (default is to move — see Phase 6).
+`--keep-inbox` skips the `tmp.old/` move (default is to move — see Phase 6) ·
+`--worktree` opts into an isolated worktree (see below; off by default).
+
+## Where the work happens — the checked-out branch, unless asked otherwise
+
+```
+RUN IN THE CURRENT BRANCH. NEVER CREATE A BRANCH, A WORKTREE, OR A PR
+ON THIS COMMAND'S OWN AUTHORITY. A WORKTREE ONLY ON AN EXPLICIT ASK.
+```
+
+Default: the branch already checked out. The roadmaps, the findings, and the
+`agents/tmp.old/` move land there as ordinary uncommitted changes for the
+operator to review. This command authorizes **analysis and authoring** — not a
+git shape ([`scope-control`](../../../../rules/scope-control.md) § Git
+operations: branch and worktree creation are permission-gated, spike and
+throwaway branches included).
+
+An isolated worktree runs only when the invocation asks for one — `--worktree`,
+or the operator saying so in the prompt. Then follow
+[`using-git-worktrees`](../../../../skills/using-git-worktrees/SKILL.md),
+including its seeding allow/deny list.
+
+Why the current branch is the right default here: this command is read-heavy and
+its whole output is a few markdown files plus a rename. Nothing it writes is a
+generated tree, so there is no build to isolate — while a worktree costs a fresh
+`node_modules`, a copied `.augment`, and a family of stale-projection failures
+that only exist because the checkout is a second one. Isolation earns its price
+when a run churns generated output or must keep a large diff off the current
+branch; an inbox triage does neither.
 
 ## The Iron Law
 
@@ -188,3 +216,6 @@ each one is and why it is spent, and let the user remove it.
   execution authorization
   ([`scope-control`](../../../../rules/scope-control.md) § Authoring vs.
   implementation).
+- Spin up a branch, a worktree, or a PR because the run feels large. Scope is
+  the operator's call; without `--worktree` or an explicit ask, the work stays
+  in the checked-out branch.

package/dist/agent-src/commands/review/changes.md

@@ -136,11 +136,12 @@ Read `subagents.adversarial_council` from `.agent-settings.yml` (default
 
 - `off` (default): skip silently.
 - `ask`: when the diff is explicitly high-risk (security, tenant, migration,
-  public API), offer it (in the user's language):
-
-  > 1. Run the adversarial verification council on this high-risk diff? (distinct-model skeptics red-team for finding coverage)
-  > 2. Skip — the five internal judges are enough
-
+  public API), offer it — in the shape
+  [`settings-ask-protocol`](../../../../rules/settings-ask-protocol.md) fixes,
+  which supplies the four slots, the one-question budget, and the storage line
+  (not persisted: `subagents.adversarial_council` is class C). What only this
+  command knows is *why now*: the diff is high-risk, and the offer is a paid
+  distinct-model red-team for finding coverage against the five internal judges.
 - `on`: run it automatically on a high-risk diff.
 
 When run, dispatch the

package/dist/agent-src/commands/roadmap/next.md

@@ -0,0 +1,194 @@
+---
+model_tier: medium
+name: roadmap-next
+pack: product-basic
+tier: 2
+visibility: internal
+cluster: roadmap
+sub: next
+skills: [roadmap-management, agent-docs-writing, ai-council, subagent-orchestration, worktree-lifecycle, git-workflow]
+description: Pick the next executable roadmap and carry it to a reviewable PR — live remote screen, five-disqualifier feasibility pass, council on the pick, process-full, chunked commits, PR, CI fix.
+argument-hint: "[--worktree] [roadmap]"
+suggestion:
+  eligible: false
+  rationale: "Cluster sub-command — reached via its cluster head's routing or its explicit /cluster:sub name; not independently suggested (surface-consolidation)."
+workspaces:
+  - agent-config-maintainer
+packs:
+  - meta
+---
+
+# /roadmap:next
+
+Selection + delivery scope of the [`/roadmap`](../roadmap.md) cluster.
+The three `process-*` subs take a roadmap **as given** and vary the
+scope inside it. `/roadmap:next` varies the **selection**: it screens
+what is actually executable right now, picks one, pins the scope to
+`full`, and carries the run to a PR that is ready for review.
+
+One invocation = one roadmap = one branch = one PR.
+
+## Instructions
+
+### 1. Live remote screen — before any selection reasoning
+
+Selection reads live state, never memory, a dashboard count, or an
+earlier fetch (per [`direct-answers`](../../rules/direct-answers.md)
+Iron Law 2 and [`roadmap-process-loop § 1`](../../contexts/execution/roadmap-process-loop.md#1-resolve-roadmap)).
+Run as ONE block:
+
+```bash
+git fetch origin --prune
+git ls-tree --name-only origin/main agents/roadmaps/          # still active?
+git ls-tree --name-only origin/main agents/roadmaps/archive/   # already closed?
+gh pr list --state open --json number,title,headRefName        # in flight?
+```
+
+Excluded from the candidate set: `template.md`, `archive/`,
+`skipped/`, `later/`, anything with `status: draft` whose promotion
+trigger has not fired, and any roadmap whose slug matches an **open
+PR's** branch — that one is taken.
+
+A roadmap completes by an archive move **inside its merging PR**, so a
+local checkout can show a closed roadmap as open. That is why this step
+precedes the screen instead of following it.
+
+### 2. Screen the survivors — five disqualifiers
+
+For each candidate read its frontmatter, its open-step count, and every
+`### blocker:` section. Disqualify on:
+
+1. **A blocker whose resolution is a human ACTION** — install a secret,
+   click a repo setting, recruit a participant, approve spend, let a
+   date pass. Apply the action-vs-judgement split: a blocker that is a
+   **judgement call** (which option, whether to reopen a recorded
+   decision, is the evidence sufficient) is **not** a disqualifier — it
+   is council-resolvable in step 3.
+2. **Spend** — paid bench or API runs, an npm publish, a public release.
+3. **A date that must pass.**
+4. **A kernel-rule edit** — [`scope-control`](../../rules/scope-control.md)
+   demands its own PR plus a ≥24 h soak, so it cannot close in this run
+   (kernel list: [`kernel-membership`](../../../docs/contracts/kernel-membership.md)).
+5. **Phases gated on a spike verdict** that may return FINDING.
+
+Also check the **family concurrency cap** (`lint_roadmap_family_cap`) —
+adding a third `road-to-<family>-*` roadmap without archiving one reds
+CI.
+
+Partial disqualification is normal and does **not** drop a candidate:
+record which phases the blocker actually blocks. A roadmap whose blocker
+touches one deferred step still has a full working set.
+
+Delegate the per-candidate reads to parallel subagents when more than
+three candidates survive step 1 (per
+[`delegation-policy`](../../rules/delegation-policy.md)); each returns
+`{slug, open_steps, blockers[], disqualifiers[], blocked_phases[]}` and
+nothing else.
+
+### 3. Rank, then let the council settle a genuine tie
+
+```
+RANK BY DEFECT SEVERITY, NEVER BY PERCENT-DONE OR STEP COUNT.
+A ROADMAP THAT REPAIRS A LIVE, USER-VISIBLE WRONGNESS OUTRANKS
+ONE THAT MOVES MORE CHECKBOXES. PICKING BY CHECKBOX COUNT IS THE
+GOODHART MOVE THIS PACKAGE'S OWN DISCIPLINE FORBIDS.
+```
+
+- **Exactly one survivor, or one that dominates** → take it, state the
+  one-line reason inline, proceed. Do not ask
+  ([`no-cheap-questions`](../../rules/no-cheap-questions.md)).
+- **Two or more with a real trade-off** → put the pick to the AI
+  council with the constraint set stated (survivors, open-step counts,
+  blocked phases, severity read) per
+  [`ai-council`](../../skills/ai-council/SKILL.md). Check the council's
+  decisive premise against the tree before adopting it — twice recorded,
+  the load-bearing claim was falsifiable in-repo.
+- **Zero survivors** → emit the screen table (candidate · open steps ·
+  disqualifier) and stop. Suggest
+  [`/roadmap:create`](create.md) or a blocker the user can clear. Do
+  not manufacture a pick.
+
+### 4. Workspace — branch by default, worktree on request
+
+Default: a feature branch in the current checkout, named
+`feat/<roadmap-slug>` per [`commit-conventions`](../../rules/commit-conventions.md).
+Propose the name once and proceed; no naming menu.
+
+`--worktree` — or the user saying "in a worktree" — routes through
+[`/worktree:create`](../../worktree/create.md) in full, including its
+[seeding allow/deny list](../../../skills/using-git-worktrees/SKILL.md#4b-seed-the-worktree--allow--deny-list).
+Two entries decide whether the run's gate results mean anything:
+dependencies are symlinked or installed **fully**, and
+`.agent-settings.yml` is **never** copied — absent IS the CI shape.
+
+### 5. Run it
+
+Hand off to [`/roadmap:process-full`](process-full.md) and follow it
+verbatim, including its Iron Law — full is full, and its five halt
+conditions are the only stops. The execution contract from
+[`roadmap-execution-contract`](../../contexts/execution/roadmap-execution-contract.md)
+is derived once; the user's single Accept covers branch, chunked
+commits, push to that branch, and the PR.
+
+### 6. Deliver
+
+Chunked commits (the agent picks the split — never ask how, per
+[`commit-policy`](../../rules/commit-policy.md)), then
+[`/create-pr`](../../git/pr/create.md): archival sweep, preflight,
+push, PR opened **ready for review** (not draft).
+
+### 7. CI — fix locally, honour the push gate
+
+Follow [`/create-pr` § 4d](../../git/pr/create.md): wait for the
+verdict in the background, never end the turn on "CI is running",
+diagnose and fix in the working tree on red, within the N=3 budget per
+[`autonomous-execution`](../../rules/autonomous-execution.md).
+
+```
+THE LOCAL FIX IS AUTONOMOUS. THE RE-PUSH IS NOT.
+git push IS A HARD-FLOOR TRIGGER AND THE FLOOR REFUSES A STANDING
+DIRECTIVE BY CONSTRUCTION. FIX, VERIFY, THEN SAY WHAT IS READY.
+NEVER INVENT A PER-PR "STANDING MANDATE" WORKAROUND.
+```
+
+The cheapest CI-fix loop is the one that never fires: run `task
+preflight` before the first push, which runs the same
+`skill_linter --changed` gate CI does.
+
+### 8. Report
+
+Roadmap picked + the screen result that picked it · steps closed /
+total · commits · PR URL as the literal last line (per
+[`direct-answers`](../../rules/direct-answers.md) reply-close) · CI
+verdict · anything left open with its reason.
+
+## Iron Law — screened, never guessed
+
+```
+NEVER PICK A ROADMAP FROM MEMORY, A DASHBOARD COUNT, OR AN EARLIER SESSION.
+THE THREE LIVE CHECKS RUN FIRST, EVERY INVOCATION.
+A ROADMAP CLAIMED "IN FLIGHT", "MERGED", OR "DONE" WITHOUT A LIVE CHECK
+AT THE MOMENT OF THE CLAIM IS AN UNVERIFIED CLAIM.
+```
+
+## Rules
+
+- **One roadmap per invocation.** Never chain a second pick after the
+  first closes — the user re-invokes.
+- **The screen is reported, not just used.** State which candidates were
+  disqualified and on what, so the pick is auditable.
+- **A judgement-call blocker goes to the council, not to the user.**
+  A human-ACTION blocker goes to neither — it disqualifies.
+- **No merge, ever.** The PR is opened for review; merging is
+  conversational ([`non-destructive-by-default`](../../rules/non-destructive-by-default.md)).
+- All rules from [`/roadmap:process-full`](process-full.md#rules) apply
+  unchanged to the execution half.
+
+## See also
+
+- [`/roadmap`](../roadmap.md) — cluster orchestrator
+- [`/roadmap:process-full`](process-full.md) — the execution half this wraps
+- [`/roadmap:ai-council`](ai-council.md) — challenge a roadmap before running it
+- [`/worktree:create`](../../worktree/create.md) — the `--worktree` path
+- [`/create-pr`](../../git/pr/create.md) — delivery + the CI wait contract
+- [`roadmap-process-loop`](../../contexts/execution/roadmap-process-loop.md) — canonical mechanics