@event4u/agent-config 7.0.1 → 7.1.0

package/.claude-plugin/marketplace.json

@@ -6,7 +6,7 @@
   },
   "metadata": {
     "description": "Shared agent configuration — skills for AI coding tools (Claude Code, Augment, Cursor, Cline, Windsurf, Gemini CLI).",
-    "version": "7.0.1",
+    "version": "7.1.0",
     "keywords": [
       "agent-config",
       "skills",
@@ -96,6 +96,7 @@
         "./.claude-plugin/skills/feature-roadmap",
         "./.claude-plugin/skills/fix",
         "./.claude-plugin/skills/fix-ci",
+        "./.claude-plugin/skills/fix-comments",
         "./.claude-plugin/skills/fix-portability",
         "./.claude-plugin/skills/fix-pr-comments",
         "./.claude-plugin/skills/fix-refs",
@@ -235,6 +236,7 @@
         "./dist/agent-src/skills/content-funnel-design",
         "./dist/agent-src/skills/context-authoring",
         "./dist/agent-src/skills/context-document",
+        "./dist/agent-src/skills/contract-review",
         "./dist/agent-src/skills/contracts-cognition",
         "./dist/agent-src/skills/conventional-commits-writing",
         "./dist/agent-src/skills/copilot-agents-optimization",
@@ -261,6 +263,7 @@
         "./dist/agent-src/skills/discovery-interview",
         "./dist/agent-src/skills/doc-coauthoring",
         "./dist/agent-src/skills/docker",
+        "./dist/agent-src/skills/dpa-review",
         "./dist/agent-src/skills/editorial-calendar",
         "./dist/agent-src/skills/eloquent",
         "./dist/agent-src/skills/emit-tickets",
@@ -321,6 +324,8 @@
         "./dist/agent-src/skills/laravel-websocket",
         "./dist/agent-src/skills/launch-readiness",
         "./dist/agent-src/skills/learning-to-rule-or-skill",
+        "./dist/agent-src/skills/legal-intake-triage",
+        "./dist/agent-src/skills/legal-practice-profile",
         "./dist/agent-src/skills/lint-skills",
         "./dist/agent-src/skills/livewire",
         "./dist/agent-src/skills/livewire-architect",
@@ -339,6 +344,7 @@
         "./dist/agent-src/skills/module-management",
         "./dist/agent-src/skills/motion-choreographer",
         "./dist/agent-src/skills/multi-tenancy",
+        "./dist/agent-src/skills/nda-triage",
         "./dist/agent-src/skills/nextjs-patterns",
         "./dist/agent-src/skills/okr-tree-modeling",
         "./dist/agent-src/skills/onboarding-design",
@@ -386,6 +392,7 @@
         "./dist/agent-src/skills/readme-writing-package",
         "./dist/agent-src/skills/reasoning-orchestrator",
         "./dist/agent-src/skills/receiving-code-review",
+        "./dist/agent-src/skills/recursive-verification",
         "./dist/agent-src/skills/refine-prompt",
         "./dist/agent-src/skills/refine-ticket",
         "./dist/agent-src/skills/release-comms",

package/CHANGELOG.md

@@ -948,6 +948,105 @@ our recommendation order, not its support status.
 > that forces a new era split (`# Era: 7.1.x`, etc.) — see
 > [`docs/contracts/CHANGELOG-conventions.md § Era splits`](docs/contracts/CHANGELOG-conventions.md).
 
+## [7.1.0](https://github.com/event4u-app/agent-config/compare/7.0.2...7.1.0) (2026-06-24)
+
+### Features
+
+* **legal:** RDG individual-case guardrail + LEGAL_NOTICE + no-definitive-language lint ([23e76df](https://github.com/event4u-app/agent-config/commit/23e76df34c612ed9bcdace069040defda80c1226))
+* **legal:** lock open-source-forever stance + harden liability disclaimer (ADR-108); close 0.6 + 3.3 ([8d828a6](https://github.com/event4u-app/agent-config/commit/8d828a64de5d23fa1c298a78b3d67c6e70ffe4b7))
+* **legal:** lock open-source-forever stance + harden liability disclaimer (ADR-108); close 0.6 + 3.3 ([f45e4e9](https://github.com/event4u-app/agent-config/commit/f45e4e99f7cccd77efb2d62fe06cb4436bc2df9e))
+* **legal:** enforcement linter, privilege markers, eval harness, docs (Phases 1/1.5/3/5) ([cb06648](https://github.com/event4u-app/agent-config/commit/cb06648009dc12daab440a273eeb780d06f6e2fa))
+* **legal:** add governed EU/DE legal pack — floor, 5 skills, vocab (ADR-107) ([b0d531f](https://github.com/event4u-app/agent-config/commit/b0d531f2701b600c93cb1ba52f3dce50e21a6d6d))
+* **bench:** pair-capture seam for recursion arm (human-preference pretest) ([7211c04](https://github.com/event4u-app/agent-config/commit/7211c043bb712cea6c0a2e0de56ea5af79b70cce))
+* **bench:** recursion gate logic + novel-lift scorer + package-recursive arm ([e33ff7d](https://github.com/event4u-app/agent-config/commit/e33ff7d52572ea35bcb8f5b190336e83e737ab1d))
+* **recursive-verification:** add depth-bounded self-verification skill + gate ([4a0c519](https://github.com/event4u-app/agent-config/commit/4a0c519628a5244cc2859cdf484661c92be25783))
+* **subagents:** A1 rule body + A2 bundle resolver + A3 response envelope ([4a9fc9b](https://github.com/event4u-app/agent-config/commit/4a9fc9b3c44c256426b0c33f4448010e890adf13))
+* **subagents:** A1 — dedicated delegation-policy auto-trigger rule ([e60cb50](https://github.com/event4u-app/agent-config/commit/e60cb50fcfb55ccabdeab88b680ec707f8a4d122))
+* **subagents:** automatic, settings-gated subagent orchestration ([dba922c](https://github.com/event4u-app/agent-config/commit/dba922c1da2177ad8103f06fd4dce49cff95977d))
+* **py2ts:** purge dead python-parity test blocks + add permanent no-python-in-src guard ([e163468](https://github.com/event4u-app/agent-config/commit/e163468af8d2b70a7e4645db0dda719c3f761cb3))
+* **roadmaps:** robust completion→archival independent of PR flow, tracking, and vendored scripts ([a879334](https://github.com/event4u-app/agent-config/commit/a87933447797c848e0e09e5d9d0ade6d280ae794))
+* **rules:** add design-fidelity rule for provided-design adherence ([88bc654](https://github.com/event4u-app/agent-config/commit/88bc6541721a7d964d670fe80ac06385c4776762))
+* add /fix:comments command to audit and trim branch code comments ([b3127b8](https://github.com/event4u-app/agent-config/commit/b3127b86733357d8fe2a32e4b3e5fe1c15447e41))
+
+### Bug Fixes
+
+* **legal:** add 'legal' to discovery-manifest schema workspace + pack enums ([f7cc834](https://github.com/event4u-app/agent-config/commit/f7cc83415d6c7ffa7600512fba059c2513c3d4df))
+* **ai-council:** raise curl timeout to 300s + --max-time so long Anthropic generations don't ETIMEDOUT ([42048b9](https://github.com/event4u-app/agent-config/commit/42048b91a5361a9041deee172609a974d926d1da))
+* **roadmap:** commit legal-pack + dashboard + archived-roadmap link fixes ([1a6910a](https://github.com/event4u-app/agent-config/commit/1a6910a39beea7e5bfc0ff21167be1463d7bfd96))
+* **test:** satisfy noUncheckedIndexedAccess in recursion tests ([29d6d89](https://github.com/event4u-app/agent-config/commit/29d6d8920192b3ade000abf9e5c102871a3ba336))
+* **py2ts:** inbox snapshot — strip macOS /private prefix so Linux CI matches ([f19436e](https://github.com/event4u-app/agent-config/commit/f19436e4a6e1cdc6dba452330e293b4414f82125))
+* **py2ts:** typecheck — assert tuple element in workspace_secrets SCAN/OBJ_CASES ([aedf6ee](https://github.com/event4u-app/agent-config/commit/aedf6eeb98eb4e171544decf4bfc9511ad7868e2))
+* **ci:** allowlist brand/iconography/premortem bare-noun skill names ([1522311](https://github.com/event4u-app/agent-config/commit/152231156b8379ded27652c28c089806a4f33d1c))
+* **ci:** register new subagents.* keys in the settings schema ([2c817a6](https://github.com/event4u-app/agent-config/commit/2c817a631c6b2062d8e5e843c31513d932d27d6f))
+* **ci:** resolve pre-existing trunk gate drift ([8515b97](https://github.com/event4u-app/agent-config/commit/8515b9793608c077f058ac42621bf6e03160e163))
+* **schema:** default the design block so settings without it still validate ([86ee94e](https://github.com/event4u-app/agent-config/commit/86ee94e313f198fd109db7dfdd6bb93a7b6dc8c6))
+* **schema:** register design.fidelity_mode in settings parity schemas ([af56652](https://github.com/event4u-app/agent-config/commit/af5665258c9b5c02bc016d2e12bfa76befbe5d5d))
+* **council:** resolve config global-only — never search the project tree ([fcd81ff](https://github.com/event4u-app/agent-config/commit/fcd81ffe92e652db88c79ba29c05b2dc0e358850))
+
+### Documentation
+
+* **recursive-verification:** record council follow-up disposition — TERMINAL ([7d132be](https://github.com/event4u-app/agent-config/commit/7d132bec2b55c00359d1d906b7701de0317f28bf))
+* **recursive-verification:** record honest-null in benchmark.md + ADR-106 ref ([87da20f](https://github.com/event4u-app/agent-config/commit/87da20ff8217e771427153eabaf7edaec4e97669))
+* add subagent-orchestration onboarding map + sync roadmap dashboard ([01b798f](https://github.com/event4u-app/agent-config/commit/01b798f42b500b3c801292f65da90137304f9243))
+* **roadmap:** add operator-runtime-harvest roadmap ([a2a92b1](https://github.com/event4u-app/agent-config/commit/a2a92b145f630e55b440b7b4214359618bf70ac9))
+* **roadmap:** record measured remaining spawn-tail count (108) + next groups ([4426293](https://github.com/event4u-app/agent-config/commit/4426293da06dfc60ff914ab9a4bc9756675b96aa))
+* **roadmap:** archive completed auto-subagent-orchestration v2 + regenerate derived ([e8af760](https://github.com/event4u-app/agent-config/commit/e8af760c792b4b7f2a7c3d61a551d84e43904df6))
+* **roadmap:** mark work_engine cluster de-pythonized (21 files) ([0b36d5c](https://github.com/event4u-app/agent-config/commit/0b36d5c4999052f33cbc423296ecc9238a2b122c))
+* **roadmap:** mark workspace_* cluster converted (12 files, +346 tests) ([154db54](https://github.com/event4u-app/agent-config/commit/154db548375a787f9c19ed38a464db8edfd4b9cb))
+* **roadmap:** capture the py2ts conversion-tail determinism trap (per-file, not a codemod) ([439817c](https://github.com/event4u-app/agent-config/commit/439817c272216f81ed2e286990f27488f3321e64))
+* **roadmap:** auto-subagent-orchestration roadmap + follow-up ([5cdbc84](https://github.com/event4u-app/agent-config/commit/5cdbc8438b0e097c99f16ee438494ae66e234c77))
+* **roadmap:** refine py2ts tail triage (36 delete-candidates / 108 convert) ([985e3e3](https://github.com/event4u-app/agent-config/commit/985e3e35bdfb48309c2dc7259e0180a39fe71cd8))
+* **roadmap:** record py2ts parity-block purge + categorize the remaining tail ([1ac823c](https://github.com/event4u-app/agent-config/commit/1ac823ce1e927a473e6c2c973cbe2e33dfb96ffb))
+* **roadmaps:** split token-saving human-measurement track off the autonomous parent ([ac188ef](https://github.com/event4u-app/agent-config/commit/ac188ef4e7044f75baf87e393f7084dae383abe3))
+* **roadmaps:** reconcile road-to-typescript-only-scripts against merged reality ([9adf45b](https://github.com/event4u-app/agent-config/commit/9adf45b3294f38f70100feaec44867074b512a2b))
+* **rdp:** restore frontier-polish closure dispositions in the archived file ([e712620](https://github.com/event4u-app/agent-config/commit/e712620f74a83f5faa50bb8885b6a1d17117adcf))
+* **rdp:** close + archive road-to-rdp-frontier-polish (L7 no-promotion, Phase 3 cancelled) ([8945d5a](https://github.com/event4u-app/agent-config/commit/8945d5ad35b15c4b177ae368f389fb0445f0c856))
+
+### Tests
+
+* **legal:** replace Gate-2 owner requirement with a test-enforced disclaimer guarantee ([60112ac](https://github.com/event4u-app/agent-config/commit/60112ac9af01cc9f85491a0b6ccaa6bd17b4caa8))
+* **legal:** add trigger-eval sets for the 5 legal skills (5 should + 5 should-not, DE+EN) ([f451af2](https://github.com/event4u-app/agent-config/commit/f451af2a4d6ee1cfd357f69fe120a905ae26e525))
+* **py2ts:** de-pythonize 21 work_engine test files (purge parity blocks / convert pure rigs) ([2221cc8](https://github.com/event4u-app/agent-config/commit/2221cc8c268bb4d7e8f7eba59d58990f31c82dfb))
+* **py2ts:** convert the workspace_* CLI parity cluster to python-free intent tests ([6697dea](https://github.com/event4u-app/agent-config/commit/6697dea75a24f363f2053c4700d8438ec2725742))
+* **py2ts:** convert workspace_hosts parity rig to python-free intent test ([6f79864](https://github.com/event4u-app/agent-config/commit/6f798649ae2f27ca1381622025102b678f283f6d))
+* **py2ts:** purge obsolete python3-vs-tsx parity blocks from 94 mixed test files ([5e07191](https://github.com/event4u-app/agent-config/commit/5e07191639e01f74f47699e62798ce6b298cbc1f))
+* **py2ts:** convert the 4 all-parity CLI rigs to python-free intent tests ([aff202c](https://github.com/event4u-app/agent-config/commit/aff202c5eb81d2a87e9147e812a2fd53635dcae2))
+* **ci-time-ratio:** drop leaked root artifact, write relative --out under ignored test-results/ ([5152942](https://github.com/event4u-app/agent-config/commit/5152942c67711f3dfdb4b35b60c696fd633a6aaf))
+
+### Chores
+
+* **legal:** regenerate agents/index.md + docs/catalog.md for legal pack artefacts ([484d5da](https://github.com/event4u-app/agent-config/commit/484d5da21ff67dd7013919b99a03fd0de3c66c72))
+* **roadmap:** archive completed recursive-verification + add legal-pack + sync dashboard ([887fbe7](https://github.com/event4u-app/agent-config/commit/887fbe7b83be52e21d751965f3dad2152d80f1ce))
+* **roadmaps:** archive py2ts teardown evidence + superseded predecessor ([d11f6cd](https://github.com/event4u-app/agent-config/commit/d11f6cd1646c43e6bd8f210b0ac703a7059236e5))
+* **condense:** re-mark stale command hashes (pre-existing drift) ([836d2c9](https://github.com/event4u-app/agent-config/commit/836d2c9ae1987d630fdce431b0ae5fed9aeb9aca))
+* register fix-comments skill in marketplace.json ([214c5d1](https://github.com/event4u-app/agent-config/commit/214c5d147b53b65c9eef56ea0e72a2d1a9224633))
+* **roadmaps:** regenerate dashboard (frontier-polish Phase 1 = done) ([9172c18](https://github.com/event4u-app/agent-config/commit/9172c18f5c6b6ad82ca6457ddbad10069a895d31))
+
+### Other
+
+* Revert "feat(legal): lock open-source-forever stance + harden liability disclaimer (ADR-108); close 0.6 + 3.3" ([1f10272](https://github.com/event4u-app/agent-config/commit/1f1027229c841cda6110b64c59747b3607b092f6))
+
+## [7.0.2](https://github.com/event4u-app/agent-config/compare/7.0.1...7.0.2) (2026-06-22)
+
+### Bug Fixes
+
+* **roadmaps:** regen entry-guard fires under symlinked invocation ([244e4e4](https://github.com/event4u-app/agent-config/commit/244e4e45a42a6b3d4793ac192e0f912d71954ba4))
+* **rdp-eval:** harden index access for strict typecheck (noUncheckedIndexedAccess) ([041b21e](https://github.com/event4u-app/agent-config/commit/041b21e4877361b1499328216242fa6ec595ff99))
+* **rdp-eval:** satisfy eslint on the new runners ([3c418e1](https://github.com/event4u-app/agent-config/commit/3c418e16faf340b8367148f6788835e400241e4e))
+
+### Documentation
+
+* **rdp-eval:** keep-scoped L6 verdict — scope orchestrator to multi-step work ([61173ca](https://github.com/event4u-app/agent-config/commit/61173ca899442b81159e4f3efe586251dcea4459))
+
+### Tests
+
+* **rdp-eval:** capture larger-N L6 + gate-classification data ([f92aad3](https://github.com/event4u-app/agent-config/commit/f92aad32620af27c5cf4d000859adeeb9894a819))
+* **rdp-eval:** port quality+L6 eval runners to TypeScript (fetch-based) ([9c4ec60](https://github.com/event4u-app/agent-config/commit/9c4ec604ca6fbb82d796cf2d65626d381a854142))
+
+### CI
+
+* **consistency:** trigger on src/** and dist/agent-src/** paths ([6a5e5e7](https://github.com/event4u-app/agent-config/commit/6a5e5e76cea3d5f879f9fcda6e510cc73039826e))
+
 ## [7.0.1](https://github.com/event4u-app/agent-config/compare/7.0.0...7.0.1) (2026-06-22)
 
 ### Bug Fixes

package/README.md

@@ -83,10 +83,12 @@ is platform operation, not a user-work flow.)
 
 > **Creative Pack — cinematic AI video.** script → character-locked image → motion+audio prompt → provider render → stitched clip, with `AIV_DRYRUN=true` as the cost-safety default. A first-class capability inside the **content / creator** experience — no longer the package's headline. See [`/video:from-script`](.augment/commands/video/from-script.md).
 
+> **Legal Pack — not legal advice.** The EU/DE legal pack (contract/NDA/DPA review, triage) is a **research-and-drafting aid only** — it does not provide legal advice, does not replace a qualified lawyer, and must not be relied on for any concrete matter. It produces general information and general templates, never individual-case examination. Read [`LEGAL_NOTICE.md`](LEGAL_NOTICE.md) before use.
+
 <details>
 <summary><b>Catalog at a glance</b> — raw artefact counts (maintainer reference)</summary>
 
-[![Skills](https://img.shields.io/badge/Skills-251-orange?style=flat-square)](dist/agent-src/skills/) [![Rules](https://img.shields.io/badge/Rules-88-orange?style=flat-square)](dist/agent-src/rules/) [![Commands](https://img.shields.io/badge/Commands-161-orange?style=flat-square)](dist/agent-src/commands/) [![Guidelines](https://img.shields.io/badge/Guidelines-82-orange?style=flat-square)](docs/guidelines/) [![Personas](https://img.shields.io/badge/Personas-26-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-257-orange?style=flat-square)](dist/agent-src/skills/) [![Rules](https://img.shields.io/badge/Rules-91-orange?style=flat-square)](dist/agent-src/rules/) [![Commands](https://img.shields.io/badge/Commands-162-orange?style=flat-square)](dist/agent-src/commands/) [![Guidelines](https://img.shields.io/badge/Guidelines-84-orange?style=flat-square)](docs/guidelines/) [![Personas](https://img.shields.io/badge/Personas-26-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/)
 
 The headline is the **experience** (profile + packs), not the raw counts. Full catalog: [`docs/catalog.md`](docs/catalog.md).
 </details>
@@ -407,7 +409,7 @@ When a prompt matches a command's purpose ("setze ticket ABC-123 um" → `/imple
 | [**Migration**](docs/MIGRATION.md) | Per-version upgrade steps |
 | [**Showcase**](docs/showcase.md) | More examples & expected behavior |
 
-Browse content: [all 161 commands](dist/agent-src/commands/) · [skills catalog](docs/skills-catalog.md) · [full catalog](docs/catalog.md) · [`llms.txt`](llms.txt).
+Browse content: [all 162 commands](dist/agent-src/commands/) · [skills catalog](docs/skills-catalog.md) · [full catalog](docs/catalog.md) · [`llms.txt`](llms.txt).
 
 ---
 

package/dist/agent-src/commands/council/default.md

@@ -88,30 +88,39 @@ Neutrality — context-handoff).
 
 ### 2. Check the council is configured + price table fresh
 
-Council config lives in **`.ai-council.yml`**, NOT in `.agent-settings.yml`.
-The CLI (`scripts/ai_council/config.py:resolve_config_path`, per
-[ADR-093](../../../docs/decisions/ADR-093-ai-council-config-user-global.md))
+Council config is **ALWAYS user-global** and lives in **`.ai-council.yml`**,
+NOT in `.agent-settings.yml`, NOT in the project tree. The CLI
+(`scripts/ai_council/config.ts:resolve_config_path`, per
+[ADR-104](../../../docs/decisions/ADR-104-ai-council-config-global-only.md),
+superseding [ADR-093](../../../docs/decisions/ADR-093-ai-council-config-user-global.md))
 resolves it — first hit wins:
 
-1. Project-local `<project_root>/agents/settings/.ai-council.yml` (if present).
-2. User-global `~/.event4u/agent-config/settings/.ai-council.yml` — default
-   home; one global config serves every project.
+1. `$AI_COUNCIL_CONFIG` — explicit absolute path (tests / power users),
+   honoured even when absent. Explicit path, never a project search.
+2. User-global `~/.event4u/agent-config/settings/.ai-council.yml` — the one
+   and only home; this single global config serves every project.
 
-Keys are **top-level** (`enabled`, `defaults:`, `members:`). No `ai_council:`
-parent block; the legacy `ai_council.*` block under `.agent-settings.yml` was
-removed in ADR-093 — never read or recommend that path for council config.
+No project-local lookup. The council never reads
+`<project_root>/agents/settings/.ai-council.yml`, `scripts/ai_council`, or any
+project file. Keys are **top-level** (`enabled`, `defaults:`, `members:`). No
+`ai_council:` parent block; the legacy `ai_council.*` block under
+`.agent-settings.yml` was removed in ADR-093 — never read or recommend that
+path for council config.
 
 ```
-NEVER claim the council "needs `.agent-settings.yml` set up" or an
-`ai_council` block. The CLI's `resolve_config_path` is authoritative:
-run `council:estimate` (Step 3) and read its exit code + message before
-making any "not configured" claim. A user-global `.ai-council.yml` with
-`enabled: true` works from EVERY project, including consumer repos that
-have no project-local config.
+THE COUNCIL CONFIG IS ALWAYS GLOBAL. NEVER SEARCH THE PROJECT FOR IT.
+NEVER claim the council "is not configured in this project". Never read
+project files — not `.agent-settings.yml`, not `scripts/ai_council` — to
+decide that. The ABSENCE OF ANY COUNCIL FILE IN THE PROJECT SAYS NOTHING
+about whether the council is configured. The CLI's `resolve_config_path`
+is authoritative: run `council:estimate` (Step 3) and read its exit code
++ message before making ANY "not configured" claim. A user-global
+`.ai-council.yml` with `enabled: true` works from EVERY project —
+consumer repos, fresh worktrees, any CWD — with zero per-project setup.
 ```
 
 CLI is the single source of truth for the configured-check — do not hand-parse
-the YAML to second-guess it. Map its outcomes (Step 4 exit codes + the
+the YAML, and never eyeball the project tree, to second-guess it. Map its outcomes (Step 4 exit codes + the
 **CLI exits 2** cases below):
 
 - `enabled: false` (top-level in resolved `.ai-council.yml`) → CLI exits 2 with
@@ -278,10 +287,9 @@ council can act on the project directly.
 ## Failure modes
 
 - **CLI exits 2, "ai_council.enabled is false"** → tell the user to set
-  top-level `enabled: true` in the resolved `.ai-council.yml` (user-global
-  `~/.event4u/agent-config/settings/.ai-council.yml`, or project-local
-  `<project_root>/agents/settings/.ai-council.yml`) — never in
-  `.agent-settings.yml`. Do not flip it autonomously.
+  top-level `enabled: true` in the user-global
+  `~/.event4u/agent-config/settings/.ai-council.yml` — never in a project
+  file, never in `.agent-settings.yml`. Do not flip it autonomously.
 - **CLI exits 2, "no council member has `enabled: true`"** → list the
   install commands (`./agent-config keys:install-anthropic`,
   `./agent-config keys:install-openai`) and stop.

package/dist/agent-src/commands/evals/fix.json

@@ -1,11 +1,15 @@
 {
   "command": "fix",
-  "intent": "Fix-workflow dispatcher — ci, pr-comments, refs, seeder, portability",
+  "intent": "Fix-workflow dispatcher — ci, pr-comments, refs, seeder, portability, comments",
   "cases": [
     {
       "prompt": "fix the failing CI",
       "expected": "fix"
     },
+    {
+      "prompt": "simplify the code comments in my branch",
+      "expected": "fix"
+    },
     {
       "prompt": "fix the broken cross-references",
       "expected": "fix"

package/dist/agent-src/commands/fix/comments.md

@@ -0,0 +1,101 @@
+---
+model_tier: medium
+name: fix-comments
+pack: engineering-base
+tier: 2
+visibility: internal
+cluster: fix
+sub: comments
+skills: [code-refactoring, quality-tools, git-workflow]
+description: Review the code comments touched by the current branch and simplify, shorten, or remove each one
+suggestion:
+  eligible: true
+  trigger_description: "simplify the comments in my branch, clean up code comments, remove redundant comments from my changes, trim comment noise"
+  trigger_context: "a feature branch with code changes pending review"
+workspaces:
+  - agent-config-maintainer
+packs:
+  - meta
+---
+
+# /fix comments
+## Instructions
+
+Audit every code comment that the current branch added or changed and
+apply the smallest edit that makes each one earn its place — simplify,
+shorten, or remove. **Comment text only — never touch executable code.**
+
+### 1. Gather the changed comments
+
+- Current branch: `git branch --show-current`. If it is the default
+  branch (`main` / `master`) → nothing to review, stop.
+- Resolve the base and collect the diff (mirrors `/review-changes`):
+  - `git diff --merge-base origin/main..HEAD` — committed changes vs. base
+  - `git diff` — unstaged changes on top
+- Scope the audit to **comments on lines the branch added or modified**
+  (inside the diff hunks). Do **not** audit comments in untouched files
+  or untouched regions — that is scope creep (`minimal-safe-diff`).
+
+If both diffs are empty → stop, nothing to review.
+
+### 2. Classify each comment
+
+For every comment in the changed regions, assign one verdict:
+
+| Verdict | When |
+|---|---|
+| **Remove** | Restates what the code already says (`// increment i`); commented-out code; obvious noise; stale comment that now contradicts the code; a TODO that is already done. |
+| **Shorten** | Correct but verbose — multi-line where one line is enough, ceremony, filler. |
+| **Simplify** | Right intent, unclear wording — tighten the phrasing without losing meaning. |
+| **Keep** | Explains *why* (rationale, gotcha, non-obvious constraint, link to an issue), or is load-bearing per the guard list below. |
+
+The bar: a comment earns its place only when it tells the reader
+something the code cannot. Default to **Keep** whenever intent is unclear.
+
+### 3. Never touch — guard list
+
+These look like comments but are load-bearing. Leave them exactly as-is:
+
+- **Tool directives** — PHPDoc that static analysis reads (`@var`,
+  `@param`, `@return`, `@template`, `@phpstan-*`, `@psalm-*`);
+  `// @ts-expect-error`, `// @ts-ignore`, `// eslint-disable*`;
+  `# type: ignore`, `# noqa`, `# pragma`, `# pylint:`; Rector / ECS /
+  Biome markers.
+- **License / copyright headers** and SPDX identifiers.
+- **Public-API doc blocks** that generate documentation, unless the
+  comment is plainly wrong.
+- `@deprecated`, `@see`, security notes, and any "why" note.
+- i18n / translation keys.
+
+When unsure whether a comment is load-bearing → **keep it**.
+
+### 4. Apply the edits
+
+- Edit each comment per its verdict with the smallest possible change.
+- Preserve indentation and the file's comment style; match the
+  comment's existing language.
+- Do **not** reflow, reformat, or re-indent the surrounding code
+  (`minimal-safe-diff`).
+
+### 5. Verify
+
+- If any doc-comment or annotation was edited in a typed language, run
+  the project's type-checker / linter on the changed files (see
+  `quality-tools`) to confirm nothing removed was load-bearing.
+- `git diff` — confirm the only changed lines are comment lines.
+
+### Output
+
+A summary table, then the totals:
+
+| File:line | Verdict | Before → After |
+|---|---|---|
+
+`Reviewed N · removed X · shortened Y · simplified Z · kept K.`
+
+### Rules
+
+- **Do NOT commit or push.** Only apply local edits.
+- **Comments only.** Never alter executable code — only comment text or
+  whole-comment removal.
+- When a comment's intent is unclear, keep it and note it in the summary.

package/dist/agent-src/commands/fix.md

@@ -3,8 +3,8 @@ model_tier: medium
 name: fix
 disable-model-invocation: true
 pack: engineering-base
-intent: "Fix-workflow dispatcher — ci, pr-comments, refs, seeder, portability"
-routes_to: [fix-ci, fix-pr-comments, fix-refs, fix-seeder, fix-portability]
+intent: "Fix-workflow dispatcher — ci, pr-comments, refs, seeder, portability, comments"
+routes_to: [fix-ci, fix-pr-comments, fix-refs, fix-seeder, fix-portability, fix-comments]
 replaces: []
 tier: 1
 visibility: advanced
@@ -36,6 +36,7 @@ with a single entry point + sub-command dispatch.
 | `/fix portability` | `commands/fix/portability.md` | Find and fix project-specific references in shared `.augment/` files |
 | `/fix seeder` | `commands/fix/seeder.md` | Scan seeder data files for broken FK references |
 | `/fix pr-comments` | `commands/fix/pr-comments.md` | Fix and reply to all open review comments — **bot + human, classified per comment**; dedupes by comment id + reply marker |
+| `/fix comments` | `commands/fix/comments.md` | Review the **code comments** in the current branch's diff and simplify, shorten, or remove each one (≠ `pr-comments`, which targets GitHub review threads) |
 
 Sub-command names match the locked contract in
 [`docs/contracts/command-clusters.md`](../docs/contracts/command-clusters.md).
@@ -52,6 +53,7 @@ the `auto_detect` kill-switch, rollback). Detection table:
 | CI run failing / "fix CI" / a CI log in context | `fix/ci` | HIGH |
 | Broken cross-references named (`/fix refs`, ref-check output) | `fix/refs` | HIGH |
 | PR review comments are the target (any "address review", PR # in context) | `fix/pr-comments` (it then resolves bot/human/both) | HIGH |
+| Source-code comments are the target ("simplify/clean up the comments in my branch", "trim comment noise") | `fix/comments` | HIGH |
 | Seeder / FK breakage named | `fix/seeder` | HIGH |
 | Project-specific leakage in shared package named | `fix/portability` | MEDIUM |
 | No clear signal, or ≥ 2 conflict | — | LOW → menu (interactive) / `ambiguous_routing` (CI) |
@@ -77,6 +79,7 @@ into it and removed.
    > 3. portability — purge project-specific refs from shared package
    > 4. seeder — scan seeders for broken FK references
    > 5. pr-comments — address open review comments (bot / human / both)
+   > 6. comments — simplify / shorten / remove code comments in the branch diff
 
 ## Rules
 

package/dist/agent-src/contexts/execution/auto-dispatch-classification.md

@@ -0,0 +1,65 @@
+# Auto-Dispatch Classification (v1 — deterministic)
+
+Decides whether a task is **delegable** to subagents, and to which
+`subagent-orchestration` mode. v1 is **rule-based and deterministic** — no
+per-turn LLM meta-call. Classification is the control plane for auto-dispatch;
+keeping it cheap and predictable is the point.
+
+## The Iron Law — ambiguity never spawns
+
+```
+A TASK IS DELEGABLE ONLY ON AN ENUMERATED SIGNAL BELOW.
+AMBIGUITY DEFAULTS TO ask / no-op — NEVER SPECULATIVE SPAWN.
+NO PER-TURN LLM CLASSIFICATION CALL IN v1.
+```
+
+## Delegable signals (v1)
+
+A task is classified **delegable** when **any** of these holds:
+
+1. **Declared parallel** — the skill/command in play carries
+   `parallelizable: steps | files | independent` in its frontmatter.
+   - `steps` → ordered plan → `do-in-steps`.
+   - `files` / `independent` → independent slices → `do-in-parallel`.
+2. **Ordered-plan structure** — the task is an explicit ordered plan (numbered
+   steps / a roadmap phase / a checklist) → `do-in-steps`.
+3. **Independent-slices structure** — N independent targets of the same shape
+   (e.g. "review these 5 files", "convert each adapter") → `do-in-parallel`.
+
+AND the **task-size floor** is cleared: the task's estimated size exceeds a
+minimum (trivial one-line edits never delegate — the dispatch overhead
+dwarfs the work). Below the floor → in-session.
+
+## Not delegable (in-session)
+
+- Trivial / single-step edits below the size floor.
+- Tasks with cross-step shared mutable state that cannot be sliced.
+- Anything that fails to match a signal above → **no-op** (in-session),
+  or **ask** when `subagents.auto == ask` and the shape is borderline.
+
+## Mode selection summary
+
+| Signal | Mode |
+|---|---|
+| `parallelizable: steps` / ordered plan | `do-in-steps` |
+| `parallelizable: files\|independent` / independent slices | `do-in-parallel` |
+| change needing verification (any of the above) | implementer + cross-model judge per the `subagent-orchestration` Iron Law |
+
+## v2+ (deferred, gated on Phase 6 evidence)
+
+LLM-based classification — only if the deterministic rules prove too coarse
+**and** the Phase-6 benchmark justifies the meta-call cost. It must be
+budgeted (consume part of the N=3 autonomous budget) and opt-in. Not in v1.
+
+## Reference implementation
+
+The deterministic rules are encoded in
+[`src/scripts/_lib/auto_dispatch.ts`](../../../../src/scripts/_lib/auto_dispatch.ts)
+(`classifyTask`), covered by
+[`tests/scripts/_lib_auto_dispatch.test.ts`](../../../../tests/scripts/_lib_auto_dispatch.test.ts).
+
+## Related
+
+- [`auto-orchestration-activation`](auto-orchestration-activation.md) — the enable/auto/manifest gate that runs before classification.
+- [`subagent-orchestration`](../../skills/subagent-orchestration/SKILL.md) — the modes this selects.
+- [`autonomous-execution`](../../rules/autonomous-execution.md) — the N=3 budget any LLM-classification v2 must respect.

package/dist/agent-src/contexts/execution/auto-orchestration-activation.md

@@ -0,0 +1,68 @@
+# Auto-Orchestration Activation
+
+How the `subagents.*` keys in `.agent-settings.yml` resolve into automatic
+subagent dispatch at runtime. Companion to
+[`subagent-configuration`](../subagent-configuration.md) (which covers the
+model / parallelism keys) and
+[`host-capability-manifest`](host-capability-manifest.md) (which decides
+whether dispatch is physically possible on the host).
+
+## Settings
+
+| Key | Default | Purpose |
+|---|---|---|
+| `subagents.enabled` | `true` | Global master switch. `false` = the whole layer is off — the canonical kill-switch. No auto-dispatch, no routing, everything runs in-session. |
+| `subagents.auto` | `ask` | Auto-dispatch mode: `off` \| `ask` \| `on`. |
+| `subagents.downshift` | `true` | Route delegable sub-tasks to the lowest-capable tier (Phase 2). |
+| `subagents.quota_arbitrage` | `true` | Prefer a separate-quota-pool model where the host manifest allows it (Phase 2 bonus). |
+| `subagents.model_map` | `{}` | Optional per-tier model override for downshift routing. |
+| `subagents.host_capabilities` | `{}` | Optional override of the host-capability manifest. |
+
+## Activation decision
+
+For any task, auto-dispatch is attempted **only when all hold**:
+
+```
+subagents.enabled == true
+AND subagents.auto != off
+AND host_manifest.subagent_spawn == true
+AND the task is classified delegable (see auto-dispatch-classification.md)
+```
+
+If any condition fails → **in-session execution**, no subagent. This is the
+safe path: a missing/false manifest field, a disabled switch, or an
+unclassifiable task all degrade silently to single-agent.
+
+## `auto` mode behaviour
+
+| Mode | When task is delegable | When not delegable |
+|---|---|---|
+| `off` | never dispatch (explicit command only) | in-session |
+| `ask` | ask once, then dispatch on yes (per `user-interaction`) | in-session |
+| `on` | dispatch; surface the chosen mode in one line | in-session |
+
+Under `ask` and `on`, the chosen `subagent-orchestration` mode and the
+per-subtask tiers are surfaced before/at dispatch — never silent.
+
+## Resolution order
+
+1. Read `subagents.*` once per run; cache for the run (no re-read mid-loop),
+   mirroring [`subagent-configuration § When settings change`](../subagent-configuration.md).
+2. Resolve the host-capability manifest once (cached) per
+   [`host-capability-manifest`](host-capability-manifest.md).
+3. Apply the activation decision above per task.
+
+## Safety floors are never lifted
+
+Auto-dispatch never bypasses a floor. The cross-model judge Iron Law
+(`subagent-orchestration`), `verify-before-complete`, the N=3 autonomous
+budget, `scope-control`, and `non-destructive-by-default` all apply to
+auto-delegated work exactly as they do to in-session work. `enabled: false`
+disables the layer; it never disables a floor.
+
+## Related
+
+- [`subagent-configuration`](../subagent-configuration.md) — model / parallelism keys.
+- [`host-capability-manifest`](host-capability-manifest.md) — the per-host gate.
+- [`auto-dispatch-classification`](auto-dispatch-classification.md) — what counts as delegable.
+- [`orchestration-telemetry`](orchestration-telemetry.md) — what each dispatch records.

package/dist/agent-src/contexts/execution/host-capability-manifest.md

@@ -0,0 +1,68 @@
+# Host-Capability Manifest
+
+Resolved once per session by the **agent** (not a long-running process) and
+cached for the rest of the session. Tells the orchestration layer which
+subagent primitives the current host actually exposes, so auto-dispatch never
+attempts a primitive the host cannot run.
+
+## Schema
+
+A single JSON object, `schema_version: 1`:
+
+```json
+{
+  "schema_version": 1,
+  "subagent_spawn": false,
+  "parallel_spawn": false,
+  "status_polling": false,
+  "separate_quota_pool": false
+}
+```
+
+## Resolution
+
+The agent fills the manifest from two inputs, in order:
+
+1. The agent's own knowledge of the **current host** and its documented
+   subagent primitives.
+2. An **optional** settings override `subagents.host_capabilities` in
+   `.agent-settings.yml`. Any field present there wins for that field; any
+   field omitted falls back to the safe default below.
+
+The manifest is resolved once and cached — the agent does not re-resolve it
+mid-session.
+
+## Safe default — unknown host assumes nothing
+
+```
+SAFE DEFAULT = ALL FIELDS false.
+AN UNKNOWN HOST IS ASSUMED TO HAVE NO SUBAGENT PRIMITIVE.
+A MISSING OR INVALID FIELD IS false, NEVER true.
+```
+
+This is a hard rule. When the host is unrecognized, or a field is absent /
+malformed in the override, the field resolves to `false`. The orchestration
+layer then degrades to in-session execution rather than attempting an
+unsupported primitive.
+
+## Fields
+
+| Field | Type | Gates downstream |
+|---|---|---|
+| `subagent_spawn` | bool | Any delegation at all. `false` → no subagent is ever spawned; everything runs in-session. |
+| `parallel_spawn` | bool | Concurrent dispatch of multiple subagents. `false` → delegation is serial, one subagent at a time. |
+| `status_polling` | bool | Monitoring running subagents (progress / completion checks). `false` → fire-and-collect only, no mid-run polling. |
+| `separate_quota_pool` | bool | Quota-arbitrage bonus (Phase 2) — subagents draw from a distinct quota pool than the session. `false` → assume shared quota. |
+
+A field being `true` is a **precondition**, not a mandate: `parallel_spawn:
+true` permits concurrency but the dispatch cap still applies (see
+`subagents.max_parallel` in subagent-configuration.md).
+
+## Related
+
+- [`subagent-configuration`](../subagent-configuration.md) — the `subagents.*`
+  settings keys and the model / parallelism resolution this manifest sits
+  alongside.
+- [`orchestration-telemetry`](orchestration-telemetry.md) — records what the
+  resolved manifest actually enabled per dispatch.
+  Phase-0 substrate belongs to.