feat(meta-orchestrator): F.3 — UX implementation per G §1.5 12 items - #205
Merged
Merged
Conversation
Closes Sub-wave F.3 from `.claude/orchestrator-prompts/meta-orchestrator-followup-audit/kickoff.md §1` per the binding scope in `docs/meta-factory/research-patches/2026-05-24-meta-orchestrator-refactor-f3-scope.md §1.5` (12 items, all addressed in this commit). ### G §1.5 items closed - **Item 1 (M1)** — §5 dispatch table row «R-phase, single» routed to Mode A inline (was Queue mode); aligns with `queue-mode.md §1 Triggers` («single kickoff → Mode A»). - **Item 2 (D3-MAJOR)** — §0 `disable-model-invocation: true` misattribution fixed: the flag suppresses CC's default auto-load into subagent contexts, NOT a recursive-invocation guard (subagent depth is hard-capped at 2 by CC's harness per `code.claude.com/docs/en/sub-agents.md`). - **Item 3 (M2)** — `references/plain-language-tail.md:8` «injects» wording replaced with «enforces presence via Stop hook `decision:block` mechanism». Mirror synced. - **Item 4 (m1)** — §5 added missing «R-phase, multiple sequential → Queue mode» row. - **Item 5 (m2)** — §6 stage-gate `created:>=2026-05-23` hardcoded date filter removed; `head:<branch> base:staging` is specific enough; T-MOB-B notes the corner-case discipline. - **Item 6 (m3)** — blank line added before `## See also` (Markdown convention + principle 09 header format). - **Item 7 (F3-S1)** — §5 antipatterns added: `#worker-dispatch-via-subagent` + `#commit-on-behalf-of-worker`, each with falsifier. - **Item 8 (F3-S2)** — §10 «Inline session report» expanded to specify the 3-layer structure (`## Dependency graph` ASCII tree + `## Action queue` 5-column table + `### Stage N` 1-liner blocks). SKILL.md stays under the 500-line gate by pointing the full grammar at `references/output-format.md` (kickoff §4 #10 split option). - **Item 9 (F3-S3)** — NEW `references/output-format.md` (299 lines) with grammar + 4 worked examples (Mode A single / Mode SDD / Mode B × N parallel worktrees / Mode Queue) + ASCII templates + 1-liner anti-patterns. Mirror synced (consumer-relative paths in §A see-also block). - **Item 10 (Gap-1)** — `launch-table-generator.sh detect_subwaves()` keyword filter added (`R-phase|execution|wiring|Mode [AB]|Direct Edit|SDD|Queue mode|I-phase|implementer|reviewer|sub-wave`) to avoid §1/§2 hook+dispatch tables being matched as sub-waves. Mirror synced + smoke-tested on this kickoff (8/8 sub-waves correctly detected). - **Item 11 (§1 Step 2 REPORT reconciliation)** — added clause: when maintainer-passed REPORT contradicts `gh pr list`, emit reconciliation note and trust mechanical state (REPORT is supplementary, not load-bearing; per 3-layer responsibility model + memory `feedback_no_human_verification_ai_self_verifies`). - **Item 12 (mirror sync)** — every edit propagated to consumer mirror (`skills/meta-orchestrator/`); helpers/launch-table-generator.sh byte-identical; SKILL.md mirror gains a condensed §10 carrying the same 6 required substrings; output-format.md mirror is identical content with consumer-relative §A links (markdown links replaced by plain text for project-internal refs so `check-skill-drift.sh` stays green). ### Principle 18 — new (`packages/core/principles/18-meta-orchestrator-output-format.test.ts`) Mechanical structural check: for each of the 4 surfaces (authoring SKILL.md §10, mirror SKILL.md §10, authoring output-format.md, mirror output-format.md), asserts the 6 required substrings are present (`## Dependency graph`, `↓`, `## Action queue`, `Paste в новый CC tab`, `Можно параллельно с`, `### Stage`). Includes companion paired-negative tests proving the check fails on synthetic broken §10 content (per principle 02 discipline). Slot 18 verified free (slots 01-17 occupied). ### Acceptance criteria - ✅ SKILL.md §10 expanded with 3-layer output spec (compact pointer + substring spec); SKILL.md stays under 500-line gate (499 lines). - ✅ `references/output-format.md` shipped with grammar + 4 worked examples (Mode A / SDD / Mode B × N / Queue). - ✅ Consumer mirror synced; `check-skill-drift.sh` green; principle 14 green. - ✅ Principle 12 still green (kickoff §5 AI-traps citations unaffected). - ✅ Principle 18 green (new). - ✅ Own cold-QA pre-handoff (T19): G §1.5 12 items + kickoff §4 #10 + §1.7 forward/backward verified. §1.7: forward+backward-check applied. Forward — principle 18 = REFERENCE from principle 12 precedent at packages/core/principles/12-ai-laziness-traps.test.ts:24; Argo `├── / └──` ADAPT vocabulary cited from F.1 prior-art PR #203 at docs/meta-factory/research-patches/2026-05-24-meta-orchestrator-ux-research.md:30; deterministic TypeScript substring grep with no paid LLM (no-paid-llm-in-ci.md §1); references/output-format.md carries Authoritative-for header per doc-authority-hierarchy.md §2 at .claude/skills/meta-orchestrator/references/output-format.md:3; SKILL ↔ references spec-of pointer at .claude/skills/meta-orchestrator/SKILL.md:421 per dual-implementation-discipline.md §5. Backward — principle 18 scope = SKILL.md §10 of meta-orchestrator skill only (covers both authoring + mirror copies via SURFACES array at packages/core/principles/18-meta-orchestrator-output-format.test.ts:54-77); does NOT introduce retroactive sweep across other skills; paired-negative tests at packages/core/principles/18-meta-orchestrator-output-format.test.ts:152-185 prove the check is non-tautological per principle 02. ### See also - Parent kickoff: `.claude/orchestrator-prompts/meta-orchestrator-followup-audit/kickoff.md §1` Sub-wave F.3 - G research-patch §1.5: `docs/meta-factory/research-patches/2026-05-24-meta-orchestrator-refactor-f3-scope.md` - F.1 prior-art: `docs/meta-factory/research-patches/2026-05-24-meta-orchestrator-ux-research.md` Prior-art: prior-art-evaluations.md#64 (subagent-driven-development KEEP-NARROW + REFERENCE — F.3 introduces no new capability beyond extending the existing meta-orchestrator skill per the G binding scope; no new SSOT row needed).
This was referenced May 24, 2026
artyhoo
added a commit
that referenced
this pull request
May 25, 2026
…SOT #72 (#222) Stage 5.A of umbrella meta-orchestrator-planner-completeness. Closes the L3 layer of the planner-completeness gap (kickoff scope vs merged-PR-30d overlap detection) per R-phase verdict #213 §5.1 and Stage 4 maintainer GO 2026-05-25. - `dup-detect.sh` — 48 impl LOC bash, two signals (xref + token Jaccard), no LLM. - `dup-detect.test.ts` — paired-negative coverage (9 cases) per principle 02. - SSOT row #72 — REFERENCE OpenHands `issue_duplicate_check_openhands.py`. Forward-check: pure deterministic bash + awk + grep at .claude/skills/meta-orchestrator/helpers/dup-detect.sh:1-71 — zero API-billed LLM per no-paid-llm-in-ci.md §1. REFERENCE OpenHands issue_duplicate_check_openhands.py cited at SSOT docs/meta-factory/prior-art-evaluations.md:72 per build-first-reuse-default.md §3. @cc-only-rationale header marker present per dual-implementation-discipline.md §6. Backward-check: net-new helper file (helpers/ goes 3→4 files); no existing signature changed. SSOT pure-append (rows #1-#71 untouched — verified by git diff origin/staging -- docs/meta-factory/prior-art-evaluations.md showing only additions). Dogfood gate flags meta-orchestrator-linear-autonomous as POTENTIAL_DUPE of #205 via xref signal at .claude/orchestrator-prompts/meta-orchestrator-linear-autonomous/kickoff.md:3 ("PR #205" explicit reference). T15 recursive self-application: the L3 layer of the planner-completeness umbrella finds a dupe of one of its own neighbouring umbrellas. §1.7: Forward+Backward checks applied — pure deterministic bash + awk at .claude/skills/meta-orchestrator/helpers/dup-detect.sh:1-71 (no-paid-llm §1), REFERENCE OpenHands at docs/meta-factory/prior-art-evaluations.md:72 (BFR §3), SSOT pure-append no row edited rows #1-#71, dogfood gate POTENTIAL_DUPE for #205 via xref at .claude/orchestrator-prompts/meta-orchestrator-linear-autonomous/kickoff.md:3. Prior-art: prior-art-evaluations.md#72 (OpenHands issue_duplicate_check_openhands.py — REFERENCE; verdict justified by T16 substrate + runtime + paid-LLM constraint mismatch with our kickoff-vs-PR, session-bound, deterministic bash design). See research-patches/2026-05-25-planner-completeness-prior-art.md §1 row 2 + §5.1 for the full 5-candidate L3 survey driving the verdict.
artyhoo
added a commit
that referenced
this pull request
May 25, 2026
… detection) (#220) Run `bash .claude/skills/meta-orchestrator/helpers/plan-currency-check.sh` (L2 Stage 3 detection shipped in #217) → 88 UNTRACKED-N entries surfaced between the 2026-05-22 reconciliation and origin/staging tip (#217). Map each to an existing §0 / Track row by adding the PR number to its evidence cell, or to a new §0 row for two umbrellas that landed in full since the prior snapshot. Re-run helper → 0 UNTRACKED remaining. Key changes: - Snapshot date 2026-05-22 → 2026-05-25 (header + §0). - N8 row: A-phase 🔲 → 🟡 — C1 SSOT-existence (#170), C2 kickoff T-enumeration floor (#174), C3 principle 13 §1.7 substance (#178), C4 delivery-channel marker (#177), activation #180. C5 + cost-levers remain gated on §5.3 utilisation trigger. - Track M.1 / M.4 → DONE: M.1 codified T20 via #212 (with NB note — recommendation-laziness took the T20 slot, mutation-equivalence T-bump 20→21 still pending); M.4 6 paired-negative bash-hook tests shipped #195/#196/#197/#198/#199/#200. - Two new §0 rows: Meta-orchestrator skill (Track P) — BUILD #186 + audit rounds #192/#193/#194/#201/#202 + UX refactor #203/#204/#205 + planner-completeness #213/#214/#217 + §1.7 PR-body mandate #216; Recommendation-laziness discipline — R-phase #206/#207, benchmark #210, I-phase Sub-waves A/C/D #211/#212/#215. - N7 row: + dogfood research-patch #135 / §4 demotion #166 / live-trial verified #171. N4b row: + design #136 / record #118. - Infra paragraph: PR refs for I.1 follow-ups (#121/#123/#124/#125/ #128/#130/#131/#143/#145/#146/#147/#148/#149/#172/#187), I.2 (#139/#142/#154/#175), I.3 DN-4 (#126/#132/#133/#138/#140/#152/ #159/#161/#162/#167). - Track 2.3 (channel-earliness audit) → DONE 2026-05-23 (#181); removed from "What actually remains". - Footer subsection: standalone work (#191 satellite-arch / #189 guard-liveness / #173 storm-readiness / #176 §10 port / #182 cleanup), Wave 10 follow-ups (#110/#112/#113), plan-revision history (#108/ #109/#153/#155/#157/#160/#164/#165/#168/#179/#185). Verification: - `bash .claude/skills/meta-orchestrator/helpers/plan-currency-check.sh | grep -c '^UNTRACKED'` → 0 (down from 88). - `npx markdownlint-cli docs/meta-factory/wave-sequencing-plan.md` → no violations. - `npx vitest run packages/core/skills/plan-currency-check.test.ts` → 14/14 passed. Prior-art: skipped — chore, doc reconciliation only, no new capability or rule introduced.
artyhoo
added a commit
that referenced
this pull request
May 25, 2026
…nding spec (13 items) (#224) Stage 1 deliverable per `.claude/orchestrator-prompts/meta-orchestrator-linear-autonomous/kickoff.md §2`. Audits PR #205 actual shipment + current staging-head (post-PR #209 follow-up) against the 13-item binding spec in G research-patch (`2026-05-24-meta-orchestrator-refactor-f3-scope.md §1.5`) + principle 18. ### Verdicts (13/13 rows filled with file:line evidence) - **CLEAN: 11/13** — Items 1, 2, 3, 4, 5, 6, 7, 8, 9, 11, 13. Items 1/4/5/6/7/11 are CLEAN in authoring; their mirror-side gap is captured once under Item 12. - **DELTA-RESOLVED-BY-FOLLOWUP: 1/13** — Item 10 (Gap-1 launch-table-generator). PR #205 alone shipped Option A keyword filter smoke-tested on only 1 kickoff; G §1.5 Falsifier required 3+ kickoffs. Closed by PR #209 (B5 hybrid + 4-kickoff smoke) — no Stage 2 action. - **DELTA-FIX-NEEDED: 1/13** — Item 12 (mirror sync). Spec requires «every item 1-11 propagated to consumer mirror» with escape-hatch scoped only to §10. Actual mirror omits 6 items beyond §10 (Items 1, 4, 5, 6, 7, 11). install.sh:209-211 ships the condensed mirror to consumer's `.claude/skills/meta-orchestrator/SKILL.md`, so mirror line 22's «see .claude/skills/.../SKILL.md after install» is a recursive pointer to the same condensed file. T16 problem-class check: variant does NOT solve spec problem-class (consumer-installed discipline coverage). One Stage 2 PR candidate. No DELTA-ACCEPTED-VARIATION verdicts. No item shipped opposite to spec (kickoff §6 F2 dramatic-regression escalation does not fire). ### Mechanical evidence - `npx vitest run packages/core/principles/18-meta-orchestrator-output-format.test.ts` → 7 passed (138 ms). - `diff -u .claude/skills/meta-orchestrator/helpers/launch-table-generator.sh skills/.../launch-table-generator.sh` → exit 0 (byte-identical). - output-format.md mirror diff = 18 lines (consumer-relative path adaptations on lines 3 + 290-296; structural content identical). - plain-language-tail.md mirror diff = 1 line (link unwrapped to plain text at :8). ### Recommendation Single Stage 2 PR to propagate Items 1/4/5/6/7/11 into mirror SKILL.md (Option a — preserve slim-consumer-surface intent). Alternative Option b — change install.sh to ship authoring copy — is a larger architectural change; (a) recommended. Item 10 needs no Stage 2 PR (already closed by #209). ### §1.7 forward+backward applied Forward — audit complies with no-paid-llm-in-ci.md §1 (deterministic git show + Read + diff + vitest, zero API-billed calls); build-first-reuse-default.md §1 (REUSE of established `project_channel_earliness_audit` + `project_memory_coverage_audit_kickoff` methodology, no new mechanism); phase-research-coverage.md §1.7 + §1.11 + §1.12 (forward+backward present, source-of-truth verified per file:line, recommendation backed by Item 12 evidence + T16 problem-class match); doc-authority-hierarchy.md §5 (folder-inherited Authoritative-for); reviewer-discipline.md §2 (Option a/b fork surfaced, recommendation is informational); parallel-subwave-isolation.md §1 (Mode A inline in isolated worktree); ai-laziness-traps.md §3 active traps T3/T13/T14/T15/T16/T19/T-F3-AF-A all applied per audit §6. Backward — audit introduces no new project-wide rule; Stage-1 deliverable per existing kickoff scope; no artefact silently superseded; Item 12 finding produces one bounded Stage 2 PR candidate, no scope creep. ### See also - Audit doc: `docs/meta-factory/research-patches/2026-05-25-linear-autonomous-audit.md` - Kickoff: `.claude/orchestrator-prompts/meta-orchestrator-linear-autonomous/kickoff.md §2 Stage 1` - Binding spec: `docs/meta-factory/research-patches/2026-05-24-meta-orchestrator-refactor-f3-scope.md §1.5` - PR being audited: #205 (squash merged 2026-05-24T21:08:36Z, merge commit 09c245a) - Item 10 follow-up patch: `docs/meta-factory/research-patches/2026-05-25-meta-orchestrator-f3-substance-followup.md` (PR #209) Prior-art: skipped — research-patch only, no new capability artifact (audit doc inherits folder-level Authoritative-for header per doc-authority-hierarchy.md §5; methodology REUSE from prior channel-earliness-audit + memory-coverage-audit precedents per CLAUDE.md «Build-vs-reuse invariant»).
This was referenced May 25, 2026
artyhoo
added a commit
that referenced
this pull request
Jul 4, 2026
…AGENTS.md + docplan-auditor (MVP) (#904) * feat(composition): MT S4 PR-B — first executable AI-doc demo in root AGENTS.md + docplan-auditor The flagship: a "Configuration access" region in the ROOT AGENTS.md, composed from two Convention nodes (a cargo type-aware ban + an npm syntax ban) with Enforced: lines COMPUTED from live cargo + npm RenderOutcomes — two-sided honest capability refusals (cargo ✅ / npm FF7001 and vice versa). A ratchet test byte-gates our own root doc (T15 self-application). agents/docplan-auditor.md adds the cold semantic node-grouping judgment layer (PR-blind, session-read, no paid LLM) — the semantic complement to the deterministic composition gate FF8001-8004. SSOT rows added this commit: #205 agents-lint (REFERENCE — competitor-neighbour, scope delimit), #206 AGENTS.md-as-render-target (ADOPT), #207 cold node-grouping semantic auditor (BUILD). Prior-art: prior-art-evaluations.md#206 (AGENTS.md as render-target for the executable-doc demo, ADOPT) Prior-art: prior-art-evaluations.md#207 (docplan-auditor — cold PR-blind node-grouping semantic audit, BUILD) * chore(mt): close multi-toolchain-convention-compiler umbrella — done.md (Final PR #904, MVP) Umbrella closure marker per CLAUDE.md "Umbrella closure convention": the last-stage PR (#904, Stage 4 PR-B) is the final MVP merge — Convention IR + cargo & npm-on-IR backends + shared frame + executable AI-doc composition plane, with the first live demo byte-gated into the root AGENTS.md. Prior-art: skipped — umbrella closure marker (done.md, 4 lines under .claude/orchestrator-prompts/), no code or capability * style(composition): prettier --write the PR-B shipped files (format:check gate) The demo + docplan-auditor files were committed un-formatted; the shipped-artifacts format:check gate (#531 dogfood) + the 4 fresh-install consumer smokes flag agents/docplan-auditor.md (a shipped agent). Whitespace-only reformat — the ratchet-gated root AGENTS.md region is untouched (excluded from the shipped surface, pre-existing table unclean on staging too) and the demo compose() output is byte-identical (ratchet still green). Prior-art: skipped — prettier --write on 3 shipped files, whitespace-only, no code/logic/capability change * test(install-sh): regen byte-identical baselines for new shipped agent docplan-auditor.md PR-B ships a new consumer agent .claude/agents/docplan-auditor.md; the install fingerprint baselines snapshot the shipped-payload file set, so all 8 (4 stacks × {greenfield,brownfield}) gain the docplan-auditor.md hash line. Only that one file line added per baseline — verified. Prior-art: skipped — snapshot baseline regen after a new shipped agent file, no new capability --------- Co-authored-by: t <t@t.co>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
.claude/orchestrator-prompts/meta-orchestrator-followup-audit/kickoff.md §1per G §1.5 12 items.references/output-format.md(299 lines) with 4 worked examples + new principle 18 structural test (substring + paired-negative).launch-table-generator.sh detect_subwaves()regex tightened with keyword filter; smoke-tested on this kickoff (8/8 sub-waves correctly detected).What changed (per G §1.5)
disable-model-invocationmisattribution fixed (it's an auto-load suppressor, not a depth guard)decision:block»created:>=2026-05-23filter## See also#worker-dispatch-via-subagent+#commit-on-behalf-of-workerreferences/output-format.mdhelpers/launch-table-generator.shskills/meta-orchestrator/**Test plan
npm --prefix packages/core run test -- --run principles/18→ 7 tests passed (4 surface checks + 1 cross-surface + 2 paired-negative).bash scripts/check-skill-drift.sh→ 0 broken refs, 0 missing frontmatter, PASS.npm --prefix packages/core run test -- --run principles→ 117/117 tests passed.wc -l= 499..mdfiles.bash .claude/skills/meta-orchestrator/helpers/launch-table-generator.sh meta-orchestrator-followup-auditreturns all 8 sub-waves correctly.gh pr checks 205→ 22 green checks (audit-self + framework + discipline + Socket Security + ci-success aggregate)./meta-orchestrator <umbrella>post-merge and verify output matches the new 3-layer structure.§1.7 Forward-check applied
Compliance against existing disciplines:
packages/core/principles/12-ai-laziness-traps.test.ts:24(same structural-check shape, different surface). Argo├── / └──ADAPT vocabulary cited from F.1 prior-art PR research(meta-orchestrator-ux): F.1 — prior-art on multi-stage dispatch UX #203 atdocs/meta-factory/research-patches/2026-05-24-meta-orchestrator-ux-research.md:30. No new BUILD without prior-art search.packages/core/principles/18-meta-orchestrator-output-format.test.ts:79-114. Zero API-billed LLM calls.references/output-format.mdcarries Authoritative-for header at.claude/skills/meta-orchestrator/references/output-format.md:3-4..claude/skills/meta-orchestrator/SKILL.md:421(Inline session report explicitly points atreferences/output-format.md).§1.7 Backward-check applied
Scope sweep of artefacts under principle 18's new rule:
packages/core/principles/18-meta-orchestrator-output-format.test.ts:54-77..claude/rules/dual-implementation-discipline.md §9forward-going-annotation discipline).packages/core/principles/18-meta-orchestrator-output-format.test.ts:152-185prove the check is non-tautological per principle 02 (packages/core/principles/02-paired-negative-test.test.ts).packages/core/principles/18-meta-orchestrator-output-format.test.ts:117-131(cross-surface agreement check).See also
.claude/orchestrator-prompts/meta-orchestrator-followup-audit/kickoff.md §1Sub-wave F.3docs/meta-factory/research-patches/2026-05-24-meta-orchestrator-refactor-f3-scope.mddocs/meta-factory/research-patches/2026-05-24-meta-orchestrator-ux-research.mdPrior-art: prior-art-evaluations.md#64 (subagent-driven-development KEEP-NARROW + REFERENCE — F.3 introduces no new capability beyond extending the existing meta-orchestrator skill per the G binding scope; no new SSOT row needed).