@chrono-meta/fh-gate 3.1.4 → 3.4.0

package/.claude/registry/agent_cards.json

@@ -50,7 +50,7 @@
       "file": "plugins/fh-meta/agents/hub-persona-auditor.md",
       "role": "Pre-publication audit of external-facing assets — 3+ persona simulation, 4-axis review (resonance/confusion/resistance/supplement), 3-tier revision proposals",
       "allowed_tools": ["Read", "Grep", "Glob"],
-      "invoked_by": ["hub-cc-pr-reviewer", "sim-conductor", "direct"],
+      "invoked_by": ["harness-pr-reviewer", "sim-conductor", "direct"],
       "writes": false
     },
     {

package/.claude/rules/fh_4axis_gate.md

@@ -45,7 +45,7 @@ paths:
 
 ## FH Improvement 4-Axis Auto-Gate (Self-Verification Orchestrator)
 
-**Whenever the AI modifies FH assets** (SKILL.md · **`SKILL_detail.md`** · `.claude/rules/*.md` · `knowledge/shared/rules/*.md` (relocated protocol rules — always full-gate, NOT under the knowledge carve-out) · `templates/` · `CLAUDE.md` · substantive `knowledge/` docs · substantive `docs/*.md` · `AGENTS.md` · `scripts/**/*.sh` · agent definitions (`plugins/*/agents/**/*.md` · `.claude/agents/**/*.md`) — see Substantive carve-out below),
+**Whenever the AI modifies FH assets** (SKILL.md · **`SKILL_detail.md`** · `.claude/rules/*.md` · `knowledge/shared/rules/*.md` (relocated protocol rules — always full-gate, NOT under the knowledge carve-out) · `templates/` · `CLAUDE.md` · substantive `knowledge/` docs · substantive `docs/*.md` · `AGENTS.md` · `scripts/**/*.sh` · `scripts/**/*.py` (🟥 `.py` since 2026-09-07 — shipped python under scripts/ was unclassified; 7 shipped files sat outside the gate, `fh_signal_2026-09-07_gate-python-blindspot.md`) · agent definitions (`plugins/*/agents/**/*.md` · `.claude/agents/**/*.md`) — see Substantive carve-out below),
 the 4-axis verification chain runs **automatically before the first commit** of that session.
 No user request is needed — this is a mandatory autonomous step, not a proposal.
 
@@ -87,6 +87,31 @@ FH asset modified → Axis 1 (templates/regression_guard.sh --pr {BRANCH})
       (`residency=(CLEAN|TAINTED|NOT_SCANNED)(...)`), and BLOCKS a `panel(...)` line that co-carries
       `residency=TAINTED(` or `residency=NOT_SCANNED(` — a sent payload and an unscreened/tainted
       one cannot both be true on the same line. Same fixtures file, cases `r1`–`r13`;
+      🟥 **since 2026-09-12 (`EVIDENCE_TOKEN_GRACE_DATE`, no retroactivity), a `panel(<families>)`
+      value ALSO requires an `evidence=` token in the same grounds** — a closed set of three:
+      `evidence=SHARED(<what every family read>)` · `evidence=INDEPENDENT(<what each got separately>)` ·
+      `evidence=MIXED(<which member got which>)`. e.g.
+      `crossfamily: panel(codex) — residency=CLEAN(files=3) · evidence=SHARED(same staged diff to both) · R1, 3 findings`.
+      **Why**: `arXiv:2609.10969` separated the two decorrelation axes at a fixed call budget
+      (48 templates, 2,880 scenarios) and a cross-model vote over SHARED evidence approved **62.9 %**
+      of unsafe proposals against **22.9 %** with an independent source — source effect **40.9 pp**
+      vs **11.3 pp** for model diversity. A `panel(codex, gemini)` that read the *same diff* is that
+      62.9 % arm while recording as this enum's strongest value; the token makes the distinction
+      sayable. 🟥 **`SHARED` is legal and common — it is not a failure.** What is blocked is a
+      `panel(...)` that does not say which it was, a value outside the three, a malformed/duplicated
+      token, and a vacuous body on `SHARED`/`MIXED` (those two are the values whose whole content is
+      *which* evidence was shared; `INDEPENDENT` is self-describing and is not body-checked, because
+      over-blocking the honest answer trains the override). Same fixtures file, cases `e1`–`e10`;
+      🟥 **`standpoint:`'s `tier2`+ grounds check BLOCKS since 2026-09-12**
+      (`STANDPOINT_GROUNDS_GRACE_DATE`, no retroactivity) — it printed `⚠️` and returned 0 before.
+      `tier2`/`tier2b`/`tier3` assert code RAN in the target, so the grounds must **name the command
+      and the output** (`ran \`bash scripts/x.sh\` there, output: 30/30 PASS`); if you only read
+      files the honest rung is `tier1b`, which is not grounds-checked. Same external number is the
+      reason: leaving the STRONGER axis advisory while hard-blocking the weaker one was not a
+      balance. Fixtures `scripts/test_marker_standpoint_lanes.sh` `N8`(blocks) / `N8b`(named command
+      passes) / `N8c`(tier1b exempt) / `N8d`(pre-grace still advisory). 🟥 Both changes gate the
+      **shape of the record**, never whether the run was real — §Mechanization Boundary's deliberate
+      residual is untouched;
       **recorded-by-convention, validated by nothing**: `axis2-rounds` (per-round yield vector) —
       steel-quench §Convergence Criteria consumes it, and a hook check for it was built and then
       REMOVED the same day for firing on 100% of markers. The convergence claim it supports is

package/.claude-plugin/marketplace.json

@@ -11,19 +11,19 @@
   "plugins": [
     {
       "name": "fh-meta",
-      "version": "3.1.4",
+      "version": "3.4.0",
       "description": "New in 2.2.0: BREAKING (gate): chamber step 6 now reads ACTUAL.md, not BUDGET.md — an in-flight chamber run whose actual cost sits in BUDGET.md blocks until the ACTUAL: line moves to tracks/_chamber/<slug>/ACTUAL.md (the runner prints the path). Why: BUDGET.md's pre-verdict hash IS the ordering witness, and step 6 hard-blocked until that same file changed, so every run that reached COMPLETE necessarily mutated a witnessed artifact and verify returned TAMPERED — the chamber's promotion condition was unsatisfiable by construction, not by strictness. Two roles (immutable witness / post-verdict calibration sink) had collided in one file; each was correct alone, so neither side's code showed the conflict. Also: ko-tech-writer Step 2/4-b scans are now calibration-backed (known-pair fixtures + reproducible command, shipped) — discrimination is proven, 'zero residue' is explicitly NOT; chamber lane suite 12 -> 33 including the runner x witness seam no test covered; chamber_run.sh now teaches the two-commit discipline (gate hashes and verdict hash must land in separate commits/PRs — it previously advised the opposite). New in 2.1.0: BREAKING (gate): `crossfamily: declined` in an Axes 2-3 marker now requires grounds naming a record path that RESOLVES on disk — bare `declined`, and `declined` justified by author judgment, are blocked at commit. Remedy: cite where the operator decision lives (e.g. `.. — operator declined sidecars, per knowledge/shared/rules/operational_adaptation.md`), or use `DEGRADED_PANEL_UNUSED` if a panel was reachable and you chose not to recruit it — which is what author judgment actually is. `declined` was the only enum value with no grounds requirement; a cross-family review then broke the first (vocabulary-grep) fix three ways — self-validating on the value's own token, vacuous keyword passes, and over-blocking real declinations in natural prose — so the check asserts a resolvable record instead of words. Also: standpoint axis gains `tier1b` (a STATIC read of a target repo, executed nothing) plus a decide-in-order procedure, after blind floor-tier sims graded pure cold-reads as `tier2` three rounds running; steel-quench Wave 1's sixth angle (gate-locality) gains the output-template row it never had, so a mandatory angle stops being structurally unreportable; verify-bidirectional gains category 5 (prescriptive doctrine statement); Sister Asset Protocol gains an active-adoption trigger; new resident doctrine — Mechanization Boundary, Local Execution First, Skeleton-not-Muscle, Expedition track, and this package's versioning policy. Hub meta-operations toolkit — 35 skills + 7 agents. New in 2.0.1: harness-doctor cadence hook, portability lint wired into pre-commit, branch_claim.sh claim-count-vs-tree-count warning, louder confidentiality-scan fail-open notice, fh-gate.sh missing-package.json survival, identity ① reclassified 🟢 (cross-harness adapters + relay argument channel). New in 1.4.53: `fh-codex-doctor` (npm bin) — Codex adapter drift scanner; reads the documented M1/M2/M3 skill tier map + skill/agent source and reports codex-native/adapter-required/claude-native/unclassified per unit, wired into `npm test`/`prepublishOnly` (fail-closed on unclassified Claude-native primitives). New in 1.4.49: steel-quench gains Step 0.6 Verdict-Invariance Probe (groundedness axis — a load-bearing judged gate's verdict must track behavior, not rubric phrasing; measured flip-count over cross-family paraphrases; arXiv:2605.06161 Policy Invariance anchor); multi_model_sidecar_strategy §Vendor-native harness (a model is strongest in its own vendor CLI — Claude/CC, GPT/codex, Gemini/Antigravity; a universal router degrades all of them, so it stays an autocomplete/QA sidecar, never orchestration); predelete_check.sh fail-closed rewrite; memory-hygiene A-TMA anchor. New in 1.4.48: phantom-quench + steel-quench gain external frontier anchors (arXiv:2607.02052 package-hallucination; arXiv:2607.02057 prompt-coverage-adequacy); README model-flat claim reframed from a per-release point-curve to structural invariants (operation flattens across tiers; depth tier-order fixed within a generation). New in 1.4.47: onboarding step ① surfaces the Mode D companion-store session-start load in the auto-read salience anchor (previously only in the local binding + rules, so a greeting could skip the load). New in 1.4.46: context-doctor command-output axis (route to rtk/proxy for verbose CLI stdout, complementing .claudeignore; risk-gated to token-scarce envs). New in 1.4.41: context-doctor 2026 trigger vocab (context engineering/rot/collapse) + phantom-citation hardening; hub measurement-integrity-checklist (cross-model measurement pre-flight: display-name pin/reps≥3/discriminating probe). New in 1.4.40: install-wizard queryable-wiki scaffold (INDEX + session-start read + R/W/C ingest). New in 1.4.39: auto-decorrelation (cross-family verifier sidecar recruitment) + video-ingest (capability-routed video ingestion). New in 1.4.x: verify-axis check-class taxonomy (mandatory-pass/measured/judged), no-reinvention Tier-0 inventory, 7-class failure taxonomy, Destructive-Op Gate, Wave-T (Temper), tier-floor governance, Mode D Model Notice, FC consent lane, default-Sonnet guidance. New in 1.3.0: public-surface-audit, field-harvest Mode B auto-trigger, 4-axis gate scope ext. Validated cross-CLI: Claude Code, Codex, Gemini.",
       "source": "./plugins/fh-meta"
     },
     {
       "name": "fh-commons",
-      "version": "3.1.4",
+      "version": "3.4.0",
       "description": "Project-agnostic utility skills — 5 skills (convergence-loop · deliberation · mcp-circuit-breaker · token-budget-gate · ko-tech-writer) + 1 agent (quench-challenger). Domain-independent utilities transplantable into any project.",
       "source": "./plugins/fh-commons"
     },
     {
       "name": "fh-qp",
-      "version": "3.1.4",
+      "version": "3.4.0",
       "description": "QP (Quality Platform) — the generic edition of a field QA harness's Prepare→Automation→Regression loop as an FH plugin: 4 skills (qp router · qp-plan · qp-run · qp-regress) + qp_tools.sh (target-class · adapter-probe · mask · surface-reach · mtm-check · run-verbs, typed exit codes) + a zero-domain-constant profile slot + 29 known-pair lanes. Drives web targets through the session's Playwright MCP and desktop targets through computer-use MCP (mobile deferred); calls a registered qasp typed capability when one exists (strictest-wins) — none is registered today, so the MCP fallback is the first edition. Verdict contract: a MACHINE closure requires a recorded assertion; a failed first step is attributed BLOCKED, not FAIL; surface_reach counts every TC in the denominator. Born as chamber run #18 (EMIT, 2026-09-05).",
       "source": "./plugins/fh-qp"
     }

package/AGENTS.md

@@ -32,7 +32,7 @@ operations or steel-quench.
 | `expert` | `plugins/fh-meta/agents/expert.md` | Web-grounded domain accuracy and current practice | `sim-conductor` Area E/D, paper review, direct |
 | `challenger` | `plugins/fh-meta/agents/challenger.md` | Evidence-cited adversarial evaluation | `steel-quench`, `harvest-loop`, `sim-conductor`, direct |
 | `fact-checker` | `plugins/fh-meta/agents/fact-checker.md` | Pre-recommendation duplicate and stale-fact search | Before new asset creation or recommendation |
-| `hub-persona-auditor` | `plugins/fh-meta/agents/hub-persona-auditor.md` | External-facing pre-publication persona audit | `hub-cc-pr-reviewer`, `sim-conductor`, direct |
+| `hub-persona-auditor` | `plugins/fh-meta/agents/hub-persona-auditor.md` | External-facing pre-publication persona audit | `harness-pr-reviewer`, `sim-conductor`, direct |
 | `quench-challenger` | `plugins/fh-commons/agents/quench-challenger.md` | Steel-quench attack plus concrete fix direction | `steel-quench` Wave 1, `install-doctor`, `marketplace-gate` |
 | `persona-innovator` | `plugins/fh-meta/agents/persona-innovator.md` | Naming gaps, frame proposals, frontier signals | `sim-conductor` Area A, `harvest-loop`, direct |
 
@@ -120,6 +120,12 @@ Because non-Claude runtimes do not auto-load Claude path rules, apply these rule
    require instrument suspicion.
 6. **Irreversible intent:** before publish, delete, or history rewrite, read and apply the
    Pre-Publish or Destructive-Op gate in `CLAUDE.md`. `pre-push` is only the git-side backstop.
+   **An automated verdict never clears an irreversible gate on its own, whatever its measured error
+   rate.** Surface class decides, not the number: a wrong finding on a review surface costs a reader
+   a minute, while on publish/delete/rewrite the wrong call is the whole loss. So improving a verdict
+   engine's score is not a route to promoting it onto an irreversible surface — the terminal step
+   stays a human or an explicit logged override. This matters here because a non-Claude runtime
+   reading this file is itself often the verdict engine in question.
 7. **Self-contrast on asset touch:** the trigger for the three-layer self-contrast (process ·
    engines · identities) is *touching an FH/PMH asset*, not being asked. Pick verification axes by
    failure mode — running all six every time is not the rule. Record, in the existing Axes 2–3
@@ -268,7 +274,7 @@ control; FH supplies the quality gate after goal completion.
 |---|---|---|
 | **M1 — Full** | No Claude-native dependency | `token-budget-gate`, `asset-placement-gate`, `phantom-quench`, `deep-clarify`, `convergence-loop`, `ko-tech-writer` (visual-QA steps degrade to text-only) |
 | **M2 — Partial** | Core works; native agent or slash-command steps need adaptation | `deliberation`, `steel-quench`, `harness-doctor`, `context-doctor`, `sim-conductor`, `harvest-loop` |
-| **M3 — Claude-only** | Requires a Claude hook or session-scoped dispatch | `goal-quench`, `hub-cc-pr-reviewer`, `install-wizard` |
+| **M3 — Claude-only** | Requires a Claude hook or session-scoped dispatch | `goal-quench`, `harness-pr-reviewer`, `install-wizard` |
 
 > **Detail**: See `knowledge/shared/harness-core/agents_md_runtime_details.md §Invocation-patterns`
 > — single, parallel, and wave composition examples — read when choosing a dispatch shape.

package/CATALOG.md

@@ -742,8 +742,8 @@ FH 전체 지도(2026-09-05) — 5층: ① 무엇인가(문 → 3단 공정 →
 - cross-family codex R1 21건(S2 = «막는다» 무조건 단정) → 수리 → R2 CONVERGED
 
 **File:** plugins/fh-meta/.claude-plugin/plugin.json + .claude-plugin/marketplace.json
-Hub meta operations tool bundle — 6 skills operation. harvest-loop path B generalization + verify-bidirectional path B generalization + frontier-digest path B generalization + cross-ecosystem-synergy-detection + plugin-recommender + **hub-cc-pr-reviewer** command tower gate operations rule automation (new). 2 agents (hub-persona-auditor + fact-checker). Beta operation — harness core principle *"beta + public release = practical capability obligation"* followed.
-- Decision: hub-cc-pr-reviewer skill newly created — command tower gate operations rule automation + PR lifecycle 4-run accumulated + explicit decision trigger
+Hub meta operations tool bundle — 6 skills operation. harvest-loop path B generalization + verify-bidirectional path B generalization + frontier-digest path B generalization + cross-ecosystem-synergy-detection + plugin-recommender + **harness-pr-reviewer** command tower gate operations rule automation (new). 2 agents (hub-persona-auditor + fact-checker). Beta operation — harness core principle *"beta + public release = practical capability obligation"* followed.
+- Decision: harness-pr-reviewer skill newly created — command tower gate operations rule automation + PR lifecycle 4-run accumulated + explicit decision trigger
 - Decision: plugin level v0.4.3 → v0.5.0 promoted — 6 skills operation baseline + path B generalization baseline followed
 - Decision: 3 skills path B generalization — harvest-loop + verify-bidirectional + frontier-digest / external user environment adaptation path enhanced
 - Note: audit-learnings deprecated from plugin (2026-05-xx) → transferred to hub-internal deprecated/; replaced by harvest-loop
@@ -763,8 +763,8 @@ Self-evolution pipeline — field-harvest → contention-layer → devil/innovat
 Bidirectional self-verification pattern automation — when user's precision counter-argument manifests after AI recommendation/agreement persistence, baseline update channel 6-step processing.
 - Decision: v0.5 official release — accumulated runs + mode C correction catch fully persisted
 
-### 2026-05-08 | fh-meta | hub-cc-pr-reviewer, command-tower-gate-automation, baseline-coherence-check, layer-5-self-catch
-**File:** plugins/fh-meta/skills/hub-cc-pr-reviewer/SKILL.md
+### 2026-05-08 | fh-meta | harness-pr-reviewer, command-tower-gate-automation, baseline-coherence-check, layer-5-self-catch
+**File:** plugins/fh-meta/skills/harness-pr-reviewer/SKILL.md
 Command Tower Gate operations rule automation — on PR input, auto-generates baseline coherence check 8-matrix + Layer 5 self-catch matrix + review comment attachment + admin override merge recommendation.
 - Decision: v0.1 newly created — PR lifecycle 4-run accumulated + explicit decision trigger met
 

package/CHEATSHEET.md

@@ -567,7 +567,7 @@ Claude agents feature
 | Skill | What it does | Say this |
 |---|---|---|
 | `install-wizard` | First-install onboarding (zshrc, sentinels, the FH self-gate) | "first-time setup", "run the install wizard" |
-| `hub-cc-pr-reviewer` | Reads a PR diff → 8-matrix baseline-consistency check → review comment + merge call | "review this PR", "check this diff" |
+| `harness-pr-reviewer` | Reads a PR diff → 8-matrix baseline-consistency check → review comment + merge call | "review this PR", "check this diff" |
 | `salience-splitter` | Splits an over-large SKILL.md that "does everything" into scoped files | "this skill is bloated", "SKILL.md too large", "split this skill" |
 
 ### Agents (sub-agents, dispatched — not slash commands)

package/CLAUDE.md

@@ -41,21 +41,21 @@ core invariants never melt). The nursery also **verifies what it births**: harne
 | **① Control Tower** | Coordinates all connected projects and **drives harness-ification across them** — decides *which* projects to harness and *when*, propagates harness assets to each, and feeds their synced learnings into the hub's compounding loop. The *how* (rules · gates · 6-axis) is executed via the Core Axis. Command HQ, not a passive registry. | `knowledge/shared/rules/auto_project_mapping.md` (mapping + **Full-Harness Mode**) · `harvest-loop` (compounding loop) · `templates/` (project-harness bundle) · `CATALOG.md` |
 | **② 프런티어 답습 · Frontier absorption** | 확신이 안 서는 자리에서 **책장(우리가 이미 가진 것) → 도서관(세상이 이미 만든 것)** 순으로 먼저 뒤진다. 목적은 **재발명 차단**이지 외부 인용으로 똑똑해 보이는 것이 아니다. | `knowledge/shared/harness-core/harness_frontier_diagnosis_*.md` · `knowledge/{your-org}/` |
 | **③ AI Collaboration Guide** | Accumulates and distributes best practices for token efficiency and dialogue methodology — "how to ask, delegate, and record". | `CHEATSHEET.md` · `knowledge/shared/dialogue/ai_dialogue_playbook.md` · `MEMORY.md` intent-based + associative recall (`knowledge/shared/dialogue/memory_intent_recall.md`) |
-| **Core Axis** | **Harness Engineering (How)** — the methodology and practice axis that realizes the three layers above. The 6-axis framework is the operating unit. **A harness is a means, not an end** — Field harness: "simpler over time" (complexity = warning signal). Meta-harness: *optimize*, not necessarily simplify — complexity earns its scope; red flags are orphaned, redundant, and decorative units, not complexity itself. | `harness_6axis_framework.md` · `hub_compounding_loop.md` · `claude_code_runtime_flow.md` · `plugins/*/agents/` (sub-agents) |
+| **Core Axis** | **Harness Engineering (How)** — the methodology and practice axis that realizes the three layers above. The 6-axis framework is the operating unit. **A harness is a means, not an end** — Field harness: "simpler over time" (complexity = warning signal). Meta-harness: *optimize*, not necessarily simplify — complexity earns its scope; red flags are orphaned, redundant, and decorative units, not complexity itself.<br>**And what it is a means *to* is Governance Engineering (What for)** — 🐿️ *«수치를 목표로 움직이되, 그 수치가 게이트를 열지는 않는다»*: drive the error rate toward 0.x% **and** hold the surfaces where no number buys passage. Two verbs, and the second one carries the load — drop it and the discipline decays into "lower the number and the gate opens", which §Irreversibility Gates forbids by name. Measured content, not a slogan: five review arms ran **2.7 %–13.6 %** claim-error on the same eight cases — every one usable on a review surface, **not one** usable on publish/delete/rewrite. That distance is why this axis exists. Canon: `governance_engineering_definition.md` (operator formulation 2026-09-09; the term exists in IT/data governance — what is ours is the referent, and an outbound claim of novelty owes a stated delta, never mere absence of prior use). | `harness_6axis_framework.md` · `hub_compounding_loop.md` · `claude_code_runtime_flow.md` · `plugins/*/agents/` (sub-agents) |
 
 > **번호가 둘이다 — 이 표의 ①②③ 은 «3층 미션»이고, 정체성 등급표의 ①~④·Ⓑ 는 «5대 정체성»이다.**
 > 같은 것을 다르게 세는 게 아니라 **다른 분류**이므로 번호가 어긋나도 모순이 아니다. 여기 **②**(프런티어 답습)는
 > 등급표에서 **④** 다. 🟥 등급은 이 표가 아니라 `ship_readiness_gate.md` 가 정본이고, 이름도 거기가 정본이다.
 
 
-> **3층 정본 — 공정 · 엔진 · 정체성**: FH 를 설명하는 뼈대는 세 층이고 셋의 관계가 정본으로 적혀 있다 — **3단 공정**(FH 의 모든 작업이 밟는 방법론 — 엔진을 벼릴 때도 같다: 초기 영혼 → 중간 **병렬 탈상관** 가속화(두 다이얼 — **탈상관**=사각 위험[모델 계열 ⓐ + 입장 ⓑ] · **병렬**=표면 크기. 곱하지 말고 골라라) → **마무리 6축 태우기**) → **4대 엔진**(영혼·품질게이트·질문하기·맥락유지) → **정체성**(방법론과 코어가 맞물려 나타나는 능력 — 5대는 단련된 실물과 등급을 가진 안정 정체성, 그 외는 방향·레버에 따라 나타나고 사라지는 면모; 운영자 정식화 2026-09-05). 기억용 형태는 **3단 공정 · 4대 엔진 · 5대 정체성 · 6축 검증**이나 🟥 **6축은 네 번째 층이 아니다** — 3단 공정 ③단계가 무엇으로 이루어지는지다. **Read `knowledge/shared/harness-core/fh_three_layer_canon.md`** before naming, re-scoping, or citing any of the three — it also defines the **6 verification axes** (ⓐ계열 · ⓑ입장 · ⓒ격리 그라운딩 · ⓓ3자대면 · ⓔ첫실사용 · ⓕ되돌림; §1-a 가 최초 4축, §1-a-2 가 2026-08-16 확장) that the third stage actually consists of, and states why the three are *not* a clean stack. 🟥 **축은 «얼마나 적대적인가»가 아니라 «무엇을 받았는가»로 갈린다** — 받는 것이 같으면 리뷰어를 몇 명 붙여도 같은 사각이 남는다. 🟥 **명칭 충돌 — 이 파일 안에 「4축」이 두 개다.** §FH Improvement **4-Axis Auto-Gate** 의 4축(Axis 1 회귀 · 2 적대 · 3 팬텀 · 4 매니페스트)은 **커밋 게이트**이고, 여기 6축은 **검증 축**이다. 부분적으로만 겹치고(Axis 1·4 는 ⓐ~ⓕ 에 대응이 없다) **서로 대체하지 않는다**. 그래서 6축은 「6축 게이트」가 아니라 「**6축 검증**」으로 부른다. Grade table stays canonical in `ship_readiness_gate.md`; this pointer never carries grades.
+> **3층 정본 — 공정 · 엔진 · 정체성**: FH 를 설명하는 뼈대는 세 층이고 셋의 관계가 정본으로 적혀 있다 — **3단 공정**(FH 의 모든 작업이 밟는 방법론 — 엔진을 벼릴 때도 같다: 초기 영혼 → 중간 **병렬 탈상관** 가속화(두 다이얼 — **탈상관**=사각 위험[모델 계열 ⓐ + 입장 ⓑ] · **병렬**=표면 크기. 곱하지 말고 골라라) → **마무리 6축 태우기**) → **4대 엔진**(영혼·품질게이트·질문하기·맥락유지) → **정체성**(방법론과 코어가 맞물려 나타나는 능력 — 5대는 단련된 실물과 등급을 가진 안정 정체성, 그 외는 방향·레버에 따라 나타나고 사라지는 면모; 운영자 정식화 2026-09-05). 기억용 형태는 **3단 공정 · 4대 엔진 · 5대 정체성 · 6축 검증**이나 🟥 **6축은 네 번째 층이 아니다** — 3단 공정 ③단계가 무엇으로 이루어지는지다. **Read `knowledge/shared/harness-core/fh_three_layer_canon.md`** before naming, re-scoping, or citing any of the three — it also defines the **6 verification axes** (ⓐ계열 · ⓑ입장 · ⓒ격리 그라운딩 · ⓓ3자대면 · ⓔ첫실사용 · ⓕ되돌림; §1-a 가 최초 4축, §1-a-2 가 2026-08-16 확장) that the third stage actually consists of, and states why the three are *not* a clean stack. 🟥 **축은 «얼마나 적대적인가»가 아니라 «무엇을 받았는가»로 갈린다** — 받는 것이 같으면 리뷰어를 몇 명 붙여도 같은 사각이 남는다. 🟢 **외부 수치가 생겼다(2026-09-12)**: `arXiv:2609.10969`(*Engineering Reliable Commit Gates for Agentic AI*, 2026-09-10)이 고정예산 2×2 로 **48 템플릿 · 2,880 시나리오**에서 두 축을 분리해 쟀다 — «**같은 증거를 읽는 교차-모델 투표**»는 위험 제안의 **62.9 %** 를 승인하고, «**독립 출처**»는 **22.9 %** 다. 출처 효과 **40.9 %p** vs 모델 다양성 효과 **11.3 %p** — **3.6 배**. 🟥 **그 수치를 읽고 바꿨다(같은 날)**: 강한 쪽(`standpoint:`)의 `tier2`+ grounds 검사가 **advisory 에서 차단으로** 올라갔고(`STANDPOINT_GROUNDS_GRACE_DATE=2026-09-12`, 소급 없음), `crossfamily:` 의 `panel(...)` 은 이제 **`evidence=SHARED|INDEPENDENT|MIXED`** 토큰을 요구한다 — «같은 diff 를 읽은 패널» 이 강한 값으로 기록되던 자리가 그 62.9 % 팔이다. 🟥 **닫힌 것은 여전히 형식이지 진위가 아니다** — 게임 가능성은 §Mechanization Boundary 가 사람에게 남긴 의도된 잔여다. 이 줄은 n=4+4 내부 관찰로 서 있었고 이제 외부 n=2,880 이 같은 방향을 가리킨다(숫자는 초록 직독으로 재확인, 2026-09-12). 🟥 **명칭 충돌 — 이 파일 안에 「4축」이 두 개다.** §FH Improvement **4-Axis Auto-Gate** 의 4축(Axis 1 회귀 · 2 적대 · 3 팬텀 · 4 매니페스트)은 **커밋 게이트**이고, 여기 6축은 **검증 축**이다. 부분적으로만 겹치고(Axis 1·4 는 ⓐ~ⓕ 에 대응이 없다) **서로 대체하지 않는다**. 그래서 6축은 「6축 게이트」가 아니라 「**6축 검증**」으로 부른다. Grade table stays canonical in `ship_readiness_gate.md`; this pointer never carries grades.
 
 > **자기 대조는 상시 의무 — 트리거는 발화가 아니라 «지금 FH/PMH 자산을 건드리고 있다»**
 > (운영자 결정 2026-08-09; 이 저장소든 **다른 사용자의 install 이든** 동일). §FH Improvement
 > 4-Axis Auto-Gate 와 **같은 트리거**이므로 새 트리거도 새 파일도 만들지 않는다 — 기록 자리는
 > **4축 마커의 기존 필드**(`axis2-*` · `axis3-*` · `residual`)다.
 > **마커에 반드시 남는 3줄** ① **①영혼** — 설계 *전에* 쓴 «성공 정의 / 절대 안 함»(없으면 `없음`)
-> · ② **돌린 축과 안 돌린 축을 각각 이름으로.** 마커 `axes-run` 은 **2026-08-17 부로 여섯 글자**를 요구한다 — **기호 키**(ⓐ계열 · ⓑ입장 · ⓒ격리 그라운딩 · ⓓ3자대면 · ⓔ첫실사용 · ⓕ되돌림). 그 전 날짜의 마커는 옛 **ASCII 네 글자**(a·b·c·d) 그대로다. 🟥 **두 배열은 같은 글자가 다른 축을 가리킨다** — 옛 `b`=첫실사용은 지금 **ⓔ**, 옛 `d`=되돌림은 지금 **ⓕ** 라서, 옛 줄을 그대로 옮기면 축 둘이 조용히 뒤바뀌고 아무 오류도 안 난다. **어느 배열인지는 마커 파일명의 날짜로 판별한다**(`< 2026-08-17` = 옛 4축). ⚠️ **표기법은 판별자가 아니다** — 초판이 «기호 키를 보면 6축인 줄 안다» 고 적었는데 **코퍼스 실측이 반증했다**: axes-run 보유 53건 중 기호 키가 4건인데 그중 **2건이 2026-08-10 자이면서 옛 4축 의미로 기호를 쓴다**(`ⓑ 첫실사용` · `ⓓ 되돌림` — 현 배열에선 각각 ⓔ·ⓕ), 혼용도 1건 있다. 훅은 그 셋을 안 읽으므로 커밋은 안 막지만 **감사자의 grep 은 거기서 틀린 답을 낸다**. ⓑ입장은 값을 여기 적지 않고 **`standpoint:` 자기 필드**를 가리킨다(`ⓑ=→standpoint`, 그 줄이 비면 죽은 포인터라 차단). 즉 산문 정본과 기계가 **축 개수로는 맞았고**, 남은 어긋남은 `standpoint:` 값의 **grounds** 한 칸이다. 🟥 **초판은 여기에 «값을 검증하는 코드가 아직 0줄»이라고 적었는데 그건 거짓이었다 — 2026-08-23 정정(RETRACTED).** 실측: `templates/.git-hooks/pre-commit` 의 `validate_standpoint_leg()` 는 **86줄**이고 `:1953` 에서 호출되어 `FAILED=1` 로 **커밋을 차단한다**(레인 = `scripts/test_marker_standpoint_lanes.sh`, 실재). 정확한 잔여는 「0줄」보다 훨씬 좁고, **그 구분을 접으면 안 된다**: ⓐ **enum 은 닫혀 있고 차단한다** — 함수 안 `return 1` **6개**(`standpoint:` 줄 부재 · 중복 `standpoint:` 줄 · enum 비-멤버 값 · `crossfamily:` 토큰 오염 · 근거 없는 `not-applicable` · 근거 없는 `DEGRADED_*`/`UNKNOWN`). ⓑ **`tier2`+ 의 «실행을 명명했는가» grounds 검사만 advisory** 다 — 미달이면 `⚠️` 를 찍고 `return 1` 을 **안 한다**(훅 스스로 *"Advisory by design"* 이라 적는다). 즉 이 줄은 「grounds 는 미검증」이라 적었어야 참이었다. **왜 그렇게 적혔나**: 2026-08-20 정정이 «enum 은 닫혔고 grounds 는 advisory» 로 이미 들어왔는데(§Standpoint-Execution-Evidence 포인터 줄이 그대로 적고 있다) **이 줄만 옛 서술로 남았다** — 같은 파일이 자기 자신과 어긋난 반쪽-픽스 전파경계다(`[[feedback_half_fix_propagation_boundary]]`). 🟥 **이 문단의 논지는 바뀌지 않는다** — 닫힌 것은 여전히 **«형식»이지 «진위»가 아니고**, 자평·게임 가능성은 그대로 열려 있다. 오히려 좁아진 만큼 정확해졌다. 형식 정본 = `.claude/rules/fh_4axis_gate.md §Marker axis fields`
+> · ② **돌린 축과 안 돌린 축을 각각 이름으로.** 마커 `axes-run` 은 **2026-08-17 부로 여섯 글자**를 요구한다 — **기호 키**(ⓐ계열 · ⓑ입장 · ⓒ격리 그라운딩 · ⓓ3자대면 · ⓔ첫실사용 · ⓕ되돌림). 그 전 날짜의 마커는 옛 **ASCII 네 글자**(a·b·c·d) 그대로다. 🟥 **두 배열은 같은 글자가 다른 축을 가리킨다** — 옛 `b`=첫실사용은 지금 **ⓔ**, 옛 `d`=되돌림은 지금 **ⓕ** 라서, 옛 줄을 그대로 옮기면 축 둘이 조용히 뒤바뀌고 아무 오류도 안 난다. **어느 배열인지는 마커 파일명의 날짜로 판별한다**(`< 2026-08-17` = 옛 4축). ⚠️ **표기법은 판별자가 아니다** — 초판이 «기호 키를 보면 6축인 줄 안다» 고 적었는데 **코퍼스 실측이 반증했다**: axes-run 보유 53건 중 기호 키가 4건인데 그중 **2건이 2026-08-10 자이면서 옛 4축 의미로 기호를 쓴다**(`ⓑ 첫실사용` · `ⓓ 되돌림` — 현 배열에선 각각 ⓔ·ⓕ), 혼용도 1건 있다. 훅은 그 셋을 안 읽으므로 커밋은 안 막지만 **감사자의 grep 은 거기서 틀린 답을 낸다**. ⓑ입장은 값을 여기 적지 않고 **`standpoint:` 자기 필드**를 가리킨다(`ⓑ=→standpoint`, 그 줄이 비면 죽은 포인터라 차단). 즉 산문 정본과 기계가 **축 개수로는 맞았고**, 남은 어긋남은 `standpoint:` 값의 **grounds** 한 칸이다. 🟥 **초판은 여기에 «값을 검증하는 코드가 아직 0줄»이라고 적었는데 그건 거짓이었다 — 2026-08-23 정정(RETRACTED).** 실측: `templates/.git-hooks/pre-commit` 의 `validate_standpoint_leg()` 는 **86줄**이고 `:1953` 에서 호출되어 `FAILED=1` 로 **커밋을 차단한다**(레인 = `scripts/test_marker_standpoint_lanes.sh`, 실재). 정확한 잔여는 「0줄」보다 훨씬 좁고, **그 구분을 접으면 안 된다**: ⓐ **enum 은 닫혀 있고 차단한다** — 함수 안 `return 1` **6개**(`standpoint:` 줄 부재 · 중복 `standpoint:` 줄 · enum 비-멤버 값 · `crossfamily:` 토큰 오염 · 근거 없는 `not-applicable` · 근거 없는 `DEGRADED_*`/`UNKNOWN`). ⓑ **`tier2`+ 의 «실행을 명명했는가» grounds 검사도 2026-09-12 부로 차단한다** — 그 전까지는 `⚠️` 만 찍고 `return 1` 을 안 했다(훅 스스로 *"Advisory by design"* 이라 적었다). 🟢 **바꾼 근거는 외부 실측이다**: `arXiv:2609.10969` 이 출처 축을 **40.9 %p**, 모델 축을 **11.3 %p** 로 재서(n=2,880 · 고정 호출예산) «강한 축의 grounds 를 advisory 로 두고 약한 축을 하드 차단» 이 균형이 아니게 됐다. grace date 로 소급하지 않으며(`< 2026-09-12` 마커는 종전 경고), 레인 = `test_marker_standpoint_lanes.sh` N8/N8b/N8c/N8d. 🟥 막는 것은 **기록의 형태**(명령과 출력을 명명했는가)이지 그 실행의 진위가 아니다. **왜 그렇게 적혔나**: 2026-08-20 정정이 «enum 은 닫혔고 grounds 는 advisory» 로 이미 들어왔는데(§Standpoint-Execution-Evidence 포인터 줄이 그대로 적고 있다) **이 줄만 옛 서술로 남았다** — 같은 파일이 자기 자신과 어긋난 반쪽-픽스 전파경계다(`[[feedback_half_fix_propagation_boundary]]`). 🟥 **이 문단의 논지는 바뀌지 않는다** — 닫힌 것은 여전히 **«형식»이지 «진위»가 아니고**, 자평·게임 가능성은 그대로 열려 있다. 오히려 좁아진 만큼 정확해졌다. 형식 정본 = `.claude/rules/fh_4axis_gate.md §Marker axis fields`
 > · ③ **각 축의 컨트롤과 그 생사**. 축을 «돌렸다»의 **최소 증거 = 컨트롤이 살아 있는 실행 출력**
 > 이다 — 안 고른 이유만 적은 것은 준수가 아니다.
 > **비용 경계**: 넷을 매번 다 돌리지 않는다. 실패 모드에 맞춰 **고른다**.
@@ -570,7 +570,7 @@ Every new `SKILL.md` must clear a **6-item bar** (role-duplication via `/asset-p
 
 ## FH Improvement 4-Axis Auto-Gate (Self-Verification Orchestrator)
 
-**FH 자산을 수정하면**(SKILL.md · **SKILL_detail.md** · `.claude/rules/*.md` · `knowledge/shared/rules/*.md` · `templates/` · `CLAUDE.md` · substantive `knowledge/`·`docs/*.md` · `AGENTS.md` · **`scripts/**/*.sh`** · **에이전트 정의**(`plugins/*/agents/**/*.md` · `.claude/agents/**/*.md`)) **4축 검증 체인이 그 세션 첫 커밋 전에 자동 실행된다.** 사용자 요청 불요 — 제안이 아니라 의무 단계다.
+**FH 자산을 수정하면**(SKILL.md · **SKILL_detail.md** · `.claude/rules/*.md` · `knowledge/shared/rules/*.md` · `templates/` · `CLAUDE.md` · substantive `knowledge/`·`docs/*.md` · `AGENTS.md` · **`scripts/**/*.sh`** · **`scripts/**/*.py`**(`.py` 는 2026-09-07 부터 — 그 전엔 출하 파이썬 7개가 게이트 밖이었다) · **에이전트 정의**(`plugins/*/agents/**/*.md` · `.claude/agents/**/*.md`)) **4축 검증 체인이 그 세션 첫 커밋 전에 자동 실행된다.** 사용자 요청 불요 — 제안이 아니라 의무 단계다.
 
 **기계층 — 무엇이 실제로 막는지 나눠서 말한다(2026-08-04 실측).** `git commit` 은 `templates/.git-hooks/pre-commit` 이 **하드 차단**한다: 축이 전부 PASS 할 때까지 커밋이 안 되고, 아래 상세가 로드되지 않아도 훅이 막는다 — 이 산문은 훅 위의 살리언스 층이지 유일 floor 가 아니다. 다만 **훅은 클라이언트측이고 `--no-verify` 로 우회된다**(§Integration branch 가 이미 그렇게 말한다 — 두 곳이 어긋나 보이던 것을 여기서 맞춘다). **그리고 서버측 검증엔 남은 잔여가 있다(2026-08-12 재확인 — `contexts=[]` 서술은 stale, 정정됨)**: `main` 은 `enforce_admins: true` 로 **푸시 경로**(PR 경유)를 강제하고, legacy `required_status_checks.contexts` 는 **`["validate"]`** — `validate` 잡(`.github/workflows/validate.yml`, 메타데이터·`selfcheck.sh` 배선 레인)이 실제 **필수 체크**다. ⚠️ **`validate` 는 Axis 1 이 아니다** — Axis 1(`regression-guard.yml` → `templates/regression_guard.sh`)은 여전히 필수 체크가 **아니고**, 그 워크플로의 `paths:` 필터가 `SKILL.md`·`.claude/rules/*.md`·`CLAUDE.md`·`templates/*.md` 만 보므로 이 절이 4축 대상으로 나열한 `knowledge/shared/rules/*.md`·`docs/*.md`·`AGENTS.md`·`scripts/**/*.sh`·에이전트 정의·`SKILL_detail.md` 만 바뀐 PR 에는 **Axis 1 자체가 돌지도 않는다**. `validate` 쪽 남은 갭은 `strict: false`: 그 체크는 PR 브랜치에 푸시할 때마다 재실행되지만(오픈 시점 한정이 아니다), 그 뒤 main 이 움직여도 재검증을 강제하지 않으므로 **초록으로 남아 있는 체크가 실제로 병합되는 최신 트리를 본 적이 없을 수 있다.** 즉 서버가 강제하는 건 *체크가 초록인가*지 *그 체크가 지금의 main 을 봤는가*가 아니다. Axes 2–3(마커)·Axis 4(매니페스트)는 그 파일들이 `tracks/**` 로 gitignored 라 CI 가 **구조적으로 볼 수조차 없다**. 정직한 표현은 "하드 차단"이 아니라 "**가용한 가장 강한 층**"이다. **미해결 잔여**: `strict` 를 켜는 것도, Axis 1 을 필수 체크로 걸거나 그 `paths:` 를 넓히는 것도 운영자 결정이다(막 flaky 레인을 하나 기록한 참이라, 과차단이 override 를 습관화시키는 쪽으로 기울 수 있다).
 
@@ -591,7 +591,32 @@ optimistic reading and miss it.**
 **Trigger (per changed file — grep-assisted, salience-dependent, no field hook)**: an AI-authored
 change to a **verdict/gate enum or exit code** (PASS/FAIL/BLOCK/allow/deny), an **irreversible-op**
 path (publish/delete/history-rewrite), or a **safety invariant** (floor, verdict-binding, a
-pre-push/pre-commit hook). Grep the diff for verdict-enum returns / gate exits / safety-marked
+pre-push/pre-commit hook). 🟥 **Registration is not a precondition** (operator decision
+2026-09-08). When the session is asked for a **merge / landing verdict** on a file that is gate code,
+this gate applies **whether or not** the file's project is mapped, tracked, or in a repo this hub
+knows. Two narrowings keep that from becoming «every review is a gate»: ⓐ the ask must be a
+**verdict on landing** (merge / ship / approve — a read-only explanation or a refactor question is
+not one) · ⓑ «gate code» is decided by **`bash scripts/gate_shape_scan.sh <file>`** — word-bounded
+verdict identifiers (a **closed** list — allow/deny/permit/approve·approval/verdict/permission/
+auth·authn·authz·authorize·authorization·authenticat*·auth_x; `author` and `allowance` do not match;
+uppercase `PASS|FAIL|BLOCK`), a bind/listen exposure, or an irreversible-op call; comment-led lines
+skipped; binary → `UNSCANNABLE`, and **exit 3 dominates a hit** — never a silent miss (known-pair
+lane `scripts/test_gate_shape_scan_lanes.sh`, 9 lanes incl. a revert probe).
+The task *naming* the file as gate/auth/exposure code is a **manual escalation on top**, not part
+of the classifier — a prompt is not a property of the file. ⓒ **FH-owned assets are excluded** —
+they already carry the 4-axis gate; this gate is for *field* code, mapped or not. Record surface for
+an unmapped file: `tracks/_meta/field_gate_review_<YYYY-MM-DD>_<slug>.md` (file · verdict ·
+`crossfamily:` verbatim from the enum · degrade-scan result · regression test landed-or-owed), and
+the reply links it. An **owed** regression test keeps the verdict `NOT-CONVERGED`; it does not
+converge on a promise. 🟥 **Pilot evidence, below bar, and CUE-DEPENDENT** (floor tier, blind, one
+variable — this paragraph injected): **before 0/1** — *"FH 자산도 매핑된 필드 하네스도 아니라서 …
+적용 대상이 아닙니다"*, reviewed bare; **after 2/3** ran the gate. Both arms' prompt ended with *"네가
+설치된 리뷰 절차가 있으면 그것을 따라라"*. 🟥 **Without that cue the gate fired 0/51** (same day, 17
+unmapped gate-shaped files × 3, same doctrine in the clone, task sentence only). So this paragraph is
+muscle, not skeleton: a session must be *reminded* to run its own procedure, and no hook supplies the
+reminder yet (`fh_signal_2026-09-08_gate-needs-cue.md`). Do not cite 2/3 without the cue condition.
+Registration was never the reason this gate exists; the degrade-direction blind spot is, and it
+does not check who owns the file. Grep the diff for verdict-enum returns / gate exits / safety-marked
 functions — strong-advisory trigger, so under-trigger is a named residual, not an airtight claim.
 
 **Gate (before merge, not after)**: ① **degrade-direction lint**
@@ -772,6 +797,19 @@ when the surface genuinely lacks its target (e.g. the code-security pass is N/A
 list ships no source/executable file — **grep the file list, don't assert "docs-only"**).
 *Applicable-but-tooling-down* is never not-applicable.
 
+**Surface class sets the error budget, not only the degrade direction** (operator decision 2026-09-08).
+The same automated verdict engine is usable on one surface and not on another, and the discriminator is
+what a wrong verdict costs. Measured that day across five review arms on the same eight cases: claim
+error rates ran **2.7 % – 13.6 %**, and every one of them is usable *on a review surface*, because a
+wrong finding costs a reader a minute. On publish · delete · history-rewrite there is no "costs a
+minute": the wrong call is the whole loss. 🟥 **So an automated verdict never clears an irreversible
+gate on its own, however good its measured rate** — it feeds a fail-closed gate whose terminal step
+stays a human or an explicit logged override. The corollary is the one that actually bites: **do not
+promote a verdict engine to an irreversible surface by improving its number.** The number is not the
+property that changes between the two surfaces. A precision figure also improves for free whenever the
+pipeline can *delete* claims, so on any surface it is meaningless beside the error rate of the
+deletions (`scripts/finding_verify.py` refuses to complete a run whose drops were never audited).
+
 A gate guarding an irreversible boundary that silently proceeds when its tooling is down is **fail-open**
 — by this floor's definition, not a gate. (The same reflex already ships piecewise — `mcp_tool_gating
 §unlisted → ask (fail-closed)`, corpus-grounding's fail-closed-no-generator — this section names the
@@ -801,7 +839,13 @@ before any scrub. So the audit must fire **pre-publish**, not after.
 **When this gate fires** — *before* any action that makes a repo/package **publicly visible for the
 first time**, especially one **derived from internal/company assets** (operator-IP that originated in a
 private harness): `gh repo create --public`, `gh repo edit --visibility public`, a first push to a new
-public remote, `npm publish`, `twine upload`, a private→public visibility flip.
+public remote, `npm publish`, `twine upload`, a private→public visibility flip, **and a scholarly
+deposit** — Zenodo / figshare / OSF publish, a DOI mint or new version, an arXiv `submit` / `replace`:
+**any action that mints or alters a persistent public identifier**, first-time or not. 🟥 **What the
+form shows is not what the server received** — check the deposit's draft **API**, the file **md5**, and
+the **machine-readable fields** (references · related identifiers · dates) separately from the body:
+`templates/PRE-PUBLISH-CHECKLIST.md` **Step 1b** (four items, un-hookable surface — prose is the floor;
+measured case: `claude_md_gate_details.md §Pre-Publish-Hook-Coverage`).
 
 **Required before the public action** (all must be non-LEAK/non-FAIL) — this gate is the **umbrella that
 invokes them**, not a competitor; when publish intent is detected, fire *this* gate (it then runs the chain),
@@ -928,7 +972,7 @@ Proposal format: `"I see [X]. Want me to run /[skill] to [one-line description]?
 | "pull this into FH", "reverse-harvest", "worth keeping", "harvest pattern", "field pattern" | `/field-harvest` |
 | **you installed or invoked an EXTERNAL asset (a tool, framework, or repo not ours) and ran it against something this hub owns** — `pip install`/`npm i` of an outside framework, cloning a peer repo to run it, adopting an upstream utility. Fires on the ACT, not on a keyword: the trigger is *"I reached outside because ours did not cover this"* | **Sister Asset Protocol** (`knowledge/shared/rules/sister_asset_protocol.md` §Active adoption) — record the resolution difference, list **items to import** AND **items the hub can propagate** (bidirectionality is a prohibition, not a nicety), and where there is no write access write a `tracks/_audit/proposal_*.md` so the operator can decide whether to contribute it upstream. 🟥 Missed 2026-08-16 on exactly this shape: an external red-team framework was installed, run against a field harness, found a real bypass — and was filed as a `type: reference` **tool pointer** with no sister audit at all |
 | "용광로모드", "crucible mode", "absorb this whole corpus", "throw everything in", "re-forge FH identity", "melt this down" (total-immersion absorption, not cherry-pick — esp. a whole corpus on a core FH axis, or a frontier showcase risking FOMO) | `knowledge/shared/harness-core/crucible_mode.md` (read it, run the chain: total-ingest → steel-quench/phantom-quench melt → governor identity-bonding → sim/persona reforge → field-harvest rebirth; the core invariants stay unmeltable) |
-| "review this PR", "check diff", "code review" | code diff → built-in `/code-review`·`/review` · FH-asset coherence → `/hub-cc-pr-reviewer` (role split) |
+| "review this PR", "check diff", "code review" | code diff → built-in `/code-review`·`/review` · FH-asset coherence → `/harness-pr-reviewer` (role split) |
 | "keep watching X", "poll this", "check every N minutes", recurring WATCH item | built-in `/loop` (interval runner) — pair with the WATCH list, don't hand-poll |
 | "research this deeply", "survey the literature", "comprehensive analysis", "deep research", "look this up thoroughly", "조사해줘", "리서치" (general topic research, not trend-scan) | **Deep-Research Capability Ladder** (`knowledge/shared/harness-core/deep_research_capability_ladder.md`) — route to the highest available rung: built-in `/deep-research` if present → else Claude `WebSearch`+`WebFetch` synthesis (tier-sensitive) → `/frontier-digest` only if it's AI/harness trend-scan. **Boundary («동향/trend»+«조사» in one utterance)**: decide by TOPIC, not verb — AI/harness-adjacent trend → `/frontier-digest`, anything else → the ladder (routing probe 2026-08-10 #8: the two vocabularies co-occur in real utterances). No-reinvention: FH routes, does not build a research engine. |
 | "orchestrate agents", "parallel dispatch", "combine skills", "multiple agents" | `/agent-composer` |
@@ -938,7 +982,7 @@ Proposal format: `"I see [X]. Want me to run /[skill] to [one-line description]?
 | "connect a project", "map this project", "link to hub" | `auto_project_mapping.md` (mapping) |
 | "harness-ify this project", "full harness setup", "프로젝트 하네스화", "promote to full harness" | `auto_project_mapping.md §6` (Full-Harness Mode) |
 | "check install", "verify setup", "confirm install", "install-doctor" | `/install-doctor` |
-| "publish", "make public", "make this repo public", "go public", "gh repo create --public", "flip to public", "first public push", "publish the package", "npm publish", "twine upload", **opening/updating a PR or pushing content to the public hub** (esp. company-origin) (publish intent — **proactive**, fire *before* the action; adding content to an already-public repo IS publishing that content) | **Pre-Publish Surface Gate** (see above → `/public-surface-audit` + `/marketplace-gate` Check 5 must PASS first). The commit-time half is now **hook-enforced** (mechanical confidentiality scan — see Pre-Publish Gate §Hook coverage (b)), so this proactive trigger is the salience layer over a mechanical floor. |
+| "publish", "make public", "make this repo public", "go public", "gh repo create --public", "flip to public", "first public push", "publish the package", "npm publish", "twine upload", **"zenodo", "figshare", "OSF", "DOI", "DOI mint / 발급", "new version" under a concept DOI, "arXiv submit / replace / 제출", "예치", "deposit"** (scholarly deposit — the deposit's *metadata* is the surface, not only its PDF), **opening/updating a PR or pushing content to the public hub** (esp. company-origin) (publish intent — **proactive**, fire *before* the action; adding content to an already-public repo IS publishing that content) | **Pre-Publish Surface Gate** (see above → `/public-surface-audit` + `/marketplace-gate` Check 5 must PASS first). The commit-time half is now **hook-enforced** (mechanical confidentiality scan — see Pre-Publish Gate §Hook coverage (b)), so this proactive trigger is the salience layer over a mechanical floor. |
 | "delete the branch", "브랜치 삭제", "브랜치 정리", "clean up branches", "force-push", "rewrite history", "지워도 돼?" (destructive intent — **proactive**, fire *before* the action) | **Destructive-Op Gate** (see above → enumerate → recover → destroy; `templates/predelete_check.sh`) |
 | **"새 기능 검증해줘", "test this feature", "이 TC 확인해줘" — verifying the user's PRODUCT/feature (not FH itself)** | **Route to the mapped field harness first** (Cross-Project Skill Bus / registry) — the field harness owns product verification. The harness-verification rows in this table (`verify-bidirectional` · `prompt-regression` · `sim-conductor` · `pipeline-conductor`) verify the *harness*, and must not shadow a product-verification ask (a field project's *harness assets* — its skills/rules — still use those FH verification rows) |
 | "지난주에 뭐 했지", "what did we do last week", "예전에 이거 한 적 있나" (recall intent) | **CATALOG-first recall** — read `CATALOG.md`, identify candidates by tag/date, then open only those files. Never scan session files one by one |
@@ -955,8 +999,8 @@ Proposal format: `"I see [X]. Want me to run /[skill] to [one-line description]?
 | "memory feels bloated", "clean up memory", "memory too large", "memory hygiene" | `/memory-hygiene` |
 | **사람이 읽을 산출물이 나가기 직전** — README·가이드·리포트·장표·PR 본문 등 «독자가 여는» 것 (proactive; 코드가 옳아도 걸린다 — 이 행이 잡는 건 정확성이 아니라 **가독성**이다) | **독자로서 한 번 읽어라** — 첫 8줄에 결론이 있나 · 본문이 고정 템플릿에 덮이지 않나 · 마지막 인상이 무엇인가. 렌즈는 이미 있다: `/sim-conductor` A-1(`beginner` cold-read) 또는 직접 렌더해서 읽기. 🟥 **정적 검사는 「없는 것」을 잡고 「안 읽히는 것」은 못 잡는다** — 실측 2건이 독립 수렴했다(qasp 축: 지적 12건 중 스캐너 적발 0 · gstack 3자대면: 배포된 리포트 본문 3줄 vs 고정 템플릿 21줄). pre-commit 이 같은 상기를 advisory 로 낸다(차단 아님) |
 | "ready to PR", "about to push", "merge this", "PR 올려줘", FH asset changed in session | 4-axis auto-gate (see above — runs automatically, no proposal needed) |
-| **field verdict/gate/safety/irreversible code changed** in a mapped project (function returning a verdict enum / gate exit code / safety-invariant · publish/delete/history path) — **proactive, before merge** | **Field-Harness Load-Bearing Change Gate** (see above → degrade-lint → cross-family review → converge; same rigor as FH assets, applied to field code) |
-| **a diff (yours or an unattended pipeline's) alters another harness's actual behavior, gate outcome, or interaction contract** — building automation that opens PRs autonomously, touching a synced/shared-body surface, or any change whose effect crosses a harness boundary (not merely a file-class match — most self-improvement is `not-applicable` here, which is the expected common case) — **proactive, before push, never as a post-PR comment** | **Standpoint axis** (`knowledge/shared/harness-core/field_verdict_crossfamily_gate.md §7` — orthogonal to `crossfamily:`; run the diff from the TARGET harness's own repo/standpoint via `tier2`/`tier2b`/`tier3`, or record `not-applicable`/`DEGRADED_*` on the closed enum. Missed once in-session while building `scripts/frontier_digest_autopilot.sh` 2026-08-15 — mis-routed to `fh-meta:hub-cc-pr-reviewer` (same-repo self-consistency, a different lens) before the operator caught it; this row exists so the next session connects the trigger without two rounds of correction.) |
+| **field verdict/gate/safety/irreversible code changed** in a mapped project **— or a merge/landing verdict asked on any gate-shaped file, mapped or not** (function returning a verdict enum / gate exit code / safety-invariant · access-control / approval / auth / exposure boundary · publish/delete/history path; «gate-shaped» = the mechanical identifier test in §Field-Harness Load-Bearing Change Gate, not a feel) — **proactive, before merge** | **Field-Harness Load-Bearing Change Gate** (see above → degrade-lint → cross-family review → converge; same rigor as FH assets, applied to field code) |
+| **a diff (yours or an unattended pipeline's) alters another harness's actual behavior, gate outcome, or interaction contract** — building automation that opens PRs autonomously, touching a synced/shared-body surface, or any change whose effect crosses a harness boundary (not merely a file-class match — most self-improvement is `not-applicable` here, which is the expected common case) — **proactive, before push, never as a post-PR comment** | **Standpoint axis** (`knowledge/shared/harness-core/field_verdict_crossfamily_gate.md §7` — orthogonal to `crossfamily:`; run the diff from the TARGET harness's own repo/standpoint via `tier2`/`tier2b`/`tier3`, or record `not-applicable`/`DEGRADED_*` on the closed enum. Missed once in-session while building `scripts/frontier_digest_autopilot.sh` 2026-08-15 — mis-routed to `fh-meta:harness-pr-reviewer` (same-repo self-consistency, a different lens) before the operator caught it; this row exists so the next session connects the trigger without two rounds of correction.) |
 | **"진단해줘", "개선해줘", "diagnose this", "improve this harness", "check this project", "audit this project"** — said while working **in a mapped project** (not a single-file ask) | **Field-Harness Diagnostic** (see §Field-Harness Diagnostic above → compose existing checks into one ranked M/S/R list → HITL approval per item, nothing auto-fixed) |
 | **"새 프로젝트", "하네스 작성해줘", "이 프로젝트 가속화", "harness-ify this", "accelerate this project"** — an onboarding/acceleration door (returning-menu ①②③) | **Onboarding / Acceleration Autopilot** (see §Onboarding / Acceleration Autopilot above → Phase 0 auto-discover + branch → innovator-centered recommend → ranked install plan → HITL per item, non-overwriting; "끝까지 자율로" → full-autonomy under /goal-quench gate) |
 

package/README.ja.md

@@ -437,8 +437,8 @@ Claude Code は作業の複雑さでモデルを自動選択しません — こ
 | [`tracks/_contrib/`](tracks/_contrib/README.md) | **同意レーン** — 非識別化した作業セッションを共有; レポが運用者たちにまたがって複利で積み上がる |
 
 > **FH 論文**: v1.0.1 方法論 · [Zenodo](https://zenodo.org/records/22542168) (DOI
-> 10.5281/zenodo.22542168) · cs.SE companion v1.2.1、プレプリント公開 ·
-> [Zenodo](https://zenodo.org/records/22635721) (DOI 10.5281/zenodo.22635721) ·
-> [arXiv:2609.04218](https://arxiv.org/abs/2609.04218) (2026-09-07 時点で arXiv は v1) · cs.AI companion は
+> 10.5281/zenodo.22542168) · cs.SE companion v1.2.2、プレプリント公開 ·
+> [Zenodo](https://zenodo.org/records/22674575) (DOI 10.5281/zenodo.22674575) ·
+> [arXiv:2609.04218](https://arxiv.org/abs/2609.04218) (v2 は 2026-09-09 公開 — §6.7 を追加し、タイトルの主張を格下げ。Zenodo v1.2.2 も同日に同一内容で公開され、両者は一致する) · cs.AI companion は
 > 準備中。これら、独立した収束的研究、そしてそれぞれの但し書き:
 > [`docs/OUTPUT_EVIDENCE.md`](docs/OUTPUT_EVIDENCE.md)。

package/README.ko.md

@@ -425,8 +425,8 @@ Claude Code 는 작업 복잡도로 모델을 자동 선택하지 않습니다.
 | [`tracks/_contrib/`](tracks/_contrib/README.md) | **동의 레인** — 비식별화된 작업 세션 공유. 레포가 운영자들에 걸쳐 복리로 쌓임 |
 
 > **FH 논문**: v1.0.1 방법론 · [Zenodo](https://zenodo.org/records/22542168) (DOI
-> 10.5281/zenodo.22542168) · cs.SE companion v1.2.1, 프리프린트 공개 ·
-> [Zenodo](https://zenodo.org/records/22635721) (DOI 10.5281/zenodo.22635721) ·
-> [arXiv:2609.04218](https://arxiv.org/abs/2609.04218) (2026-09-07 기준 arXiv 는 v1) · cs.AI companion
+> 10.5281/zenodo.22542168) · cs.SE companion v1.2.2, 프리프린트 공개 ·
+> [Zenodo](https://zenodo.org/records/22674575) (DOI 10.5281/zenodo.22674575) ·
+> [arXiv:2609.04218](https://arxiv.org/abs/2609.04218) (v2 는 2026-09-09 등재 — §6.7 신설 · 제목의 주장 등급강등. Zenodo v1.2.2 도 같은 날 같은 내용으로 발행돼 두 예치가 일치한다) · cs.AI companion
 > 준비 중. 이것들과 독립적인 수렴 연구, 그리고 각각의 주의사항:
 > [`docs/OUTPUT_EVIDENCE.md`](docs/OUTPUT_EVIDENCE.md).

package/README.md

@@ -416,8 +416,8 @@ and the phrase that triggers it:
 | [`tracks/_contrib/`](tracks/_contrib/README.md) | **Consent lane** — share a de-identified work session; the repo compounds across operators |
 
 > **FH papers**: v1.0.1 methodology · [Zenodo](https://zenodo.org/records/22542168) (DOI
-> 10.5281/zenodo.22542168) · cs.SE companion v1.2.1, preprint ·
-> [Zenodo](https://zenodo.org/records/22635721) (DOI 10.5281/zenodo.22635721) ·
-> [arXiv:2609.04218](https://arxiv.org/abs/2609.04218) (arXiv shows v1 as of 2026-09-07) · cs.AI companion in
+> 10.5281/zenodo.22542168) · cs.SE companion v1.2.2, preprint ·
+> [Zenodo](https://zenodo.org/records/22674575) (DOI 10.5281/zenodo.22674575) ·
+> [arXiv:2609.04218](https://arxiv.org/abs/2609.04218) (v2 announced 2026-09-09 — adds Sec. 6.7 and downgrades the title's claim; Zenodo v1.2.2 was published the same day with the same content, so the two deposits agree) · cs.AI companion in
 > preparation. Those, the independent convergent work, and the caveats on each:
 > [`docs/OUTPUT_EVIDENCE.md`](docs/OUTPUT_EVIDENCE.md).

package/README.zh.md

@@ -404,8 +404,8 @@ Claude Code 不会按任务复杂度自动选择模型 —— 这个要你设置
 | [`tracks/_contrib/`](tracks/_contrib/README.md) | **同意通道** —— 分享一个去标识化的工作会话；仓库在众多操作者之间复利累积 |
 
 > **FH 论文**：v1.0.1 方法论 · [Zenodo](https://zenodo.org/records/22542168)（DOI
-> 10.5281/zenodo.22542168）· cs.SE companion v1.2.1，预印本公开 ·
-> [Zenodo](https://zenodo.org/records/22635721)（DOI 10.5281/zenodo.22635721）·
-> [arXiv:2609.04218](https://arxiv.org/abs/2609.04218)（截至 2026-09-07，arXiv 为 v1）· cs.AI companion
+> 10.5281/zenodo.22542168）· cs.SE companion v1.2.2，预印本公开 ·
+> [Zenodo](https://zenodo.org/records/22674575)（DOI 10.5281/zenodo.22674575）·
+> [arXiv:2609.04218](https://arxiv.org/abs/2609.04218)（v2 于 2026-09-09 公布 — 新增 §6.7 并下调标题主张；Zenodo v1.2.2 于同日以相同内容发布，两处存档一致）· cs.AI companion
 > 筹备中。这些、独立的收敛性工作，以及每一项的注意事项：
 > [`docs/OUTPUT_EVIDENCE.md`](docs/OUTPUT_EVIDENCE.md)。

package/docs/OUTPUT_EVIDENCE.md

@@ -36,7 +36,7 @@
 | Artifact | Reference |
 |---|---|
 | Paper v1.0.1 — methodology | Zenodo DOI [`10.5281/zenodo.22542168`](https://zenodo.org/records/22542168) (all versions: `10.5281/zenodo.20397565`) — 2-layer design, 6-axis framework, 4-agent orchestration, compounding loop, with empirical evidence. **arXiv: rejected at moderation (2026-09-06)**; v1.0.1 is the corrected version (11 of 17 reference entries in v1.0 did not match the works cited — see the erratum in the record). Do not read the rejection as an assessment of the methodology, and do not read v1.0.1 as re-reviewed: it has not been resubmitted |
-| cs.SE companion — governance-gate methodology | **preprint, publicly posted** · Zenodo DOI [`10.5281/zenodo.22635721`](https://zenodo.org/records/22635721) (v1.2.1; the record's description carries the per-version change notes — v1.2 added the replication section and withdrew two previously reported results; v1.2.1 propagates that withdrawal into the abstract and the conclusion, which v1.2 had left unrevised; all versions `10.5281/zenodo.20680080` · CC-BY-4.0) · [`arXiv:2609.04218`](https://arxiv.org/abs/2609.04218) (cs.SE; submitted 2026-07-01, announced 2026-09). 🟥 **As of 2026-09-07 the arXiv record is still v1**, whose abstract carries the severity-differential sentence that §6.6 of the linked Zenodo record withdraws — open both and compare. **Author-reported, not visible on the public record:** a v2 with the v1.2.1 content was submitted 2026-09-07 and its submission confirmation gives a *scheduled* announcement date of 2026-09-09. 🟥 arXiv is a moderated preprint server; moderation is not peer review, and neither the Zenodo deposit nor the arXiv posting establishes peer review, editorial acceptance, or venue acceptance |
+| cs.SE companion — governance-gate methodology | **preprint, publicly posted** · Zenodo DOI [`10.5281/zenodo.22674575`](https://zenodo.org/records/22674575) (v1.2.2; the record's description carries the per-version change notes — v1.2 added the replication section and withdrew two previously reported results; v1.2.1 propagates that withdrawal into the abstract and the conclusion, which v1.2 had left unrevised; v1.2.2 adds Sec. 6.7 and downgrades the title and abstract accordingly; all versions `10.5281/zenodo.20680080` · CC-BY-4.0) · [`arXiv:2609.04218`](https://arxiv.org/abs/2609.04218) (cs.SE; submitted 2026-07-01, announced 2026-09). **v2 announced 2026-09-09** and is on the public record: it carries the v1.2.2 content, adds §6.7 (an independent-session re-test of ground-truth recall under condition-blind scoring, which **does not replicate** §6.2's strict-recall contrast — the two conditions come out one strict hit apart in 24), downgrades the title's last clause from *Evidence* to *a Test of*, and turns the abstract's closing claim into a **hypothesis**. ✅ **The two deposits agree as of 2026-09-09.** Zenodo v1.2.2 (DOI `10.5281/zenodo.22674575`) was published that day carrying the same content as arXiv v2, and the record declares a machine-readable `isIdenticalTo arXiv:2609.04218` relation. This was verified by reading the **published record's API**, not the submission form — creators, license, keywords, the description, the file md5, and the version count were re-read after publication and after the metadata edit, and none changed. The earlier divergence (arXiv v1.2.2 against Zenodo v1.2.1) is therefore closed; the superseded v1.2.1 remains separately citable at `10.5281/zenodo.22635721`. 🟥 arXiv is a moderated preprint server; moderation is not peer review, and neither the Zenodo deposit nor the arXiv posting establishes peer review, editorial acceptance, or venue acceptance |
 | cs.AI companion — "Governance Dividend" | in preparation |
 | Package | npm [`@chrono-meta/fh-gate`](https://www.npmjs.com/package/@chrono-meta/fh-gate) — multi-backend governance gate (claude · codex · auto) |
 | Codex-compatible | `docs/codex-compat.md` — methodology layer runs model-agnostic. Marked **beta** there in the *validation-maturity* sense (external validation is still thin), **not** the *scope* sense: partial automation-layer support is the design, not an unfinished state |
@@ -131,7 +131,7 @@ rather than only synthetic ones.
 
 ```bash
 # active skills. NOTE: the old recipe here grepped each SKILL.md for "redirect stub"/"deprecated"
-# and returned 38, because phantom-quench and hub-cc-pr-reviewer — both live — merely MENTION those
+# and returned 38, because phantom-quench and harness-pr-reviewer — both live — merely MENTION those
 # words in their prose. A body-text grep cannot tell "I am a stub" from "I detect stubs". There are
 # currently zero stubs, so count the files and re-introduce an exclusion only when one exists, in
 # frontmatter where it can be matched on a field rather than on a phrase.

package/docs/STANDARDS_ALIGNMENT.md

@@ -41,7 +41,7 @@ to FH itself and to the field harnesses it emits.
 | 42001 A.7 data (A.7.2–A.7.6) | acquisition, quality, provenance, preparation | FH holds no training data but does hold **evaluation data** (probes, corpora, transcripts, logs): provenance = sim header `corpus_head_date` · quality = known-pair calibration · preparation = residency strip · **company residency is absolute** | `probes_live.yaml`, `scripts/residency_closure_scan.py` | 🟡 provenance/preparation present; no written acquisition or quality criteria for evaluation data |
 | ISO/IEC 5338 life cycle | define · control · execute · improve | incubator (chamber) → EMIT → field → harvest-loop compounding | `harness_incubator_doctrine.md`, `hub_compounding_loop.md` | ✅ |
 | ISO/IEC 23894 risk management | identify · analyse · evaluate · treat | ship-readiness grades · Surface-Class Degrade Invariant (irreversible = fail-closed) | `ship_readiness_gate.md`, `CLAUDE.md §Irreversibility Gates` | ✅ partial |
-| ISO/IEC 20246 reviews | work-product reviews | cross-family review · `/apex-review` · `hub-cc-pr-reviewer` · cold read (`beginner`) | `plugins/fh-meta/agents/` | ✅ |
+| ISO/IEC 20246 reviews | work-product reviews | cross-family review · `/apex-review` · `harness-pr-reviewer` · cold read (`beginner`) | `plugins/fh-meta/agents/` | ✅ |
 
 ## What a field harness inherits
 A harness FH emits gets the same rows for free where it adopts the templates (`templates/.git-hooks/*`, the marker schema, the

package/docs/codex-compat.md

@@ -113,7 +113,7 @@ When `codex exec` runs **inside this repo**, FH's Claude-native git/Stop/PostToo
 M2 skills (`deliberation`, `steel-quench`, `harness-doctor`, `context-doctor`, `sim-conductor`, `harvest-loop`) have a core workflow that runs under Codex, but any step that dispatches `Agent(subagent_type=...)` or a slash command must be replaced by `fh-run` or a direct `codex exec` call reading the sub-agent's `SKILL.md`/agent `.md` — same workflow, different runtime (the "M2 adaptation pattern" in `AGENTS.md`). Example: `steel-quench` Waves 1–3 run; the `quench-challenger` agent step becomes `fh-run --agent fh-commons:quench-challenger`.
 
 ### 3. M3 skills do not run automatically under Codex
-M3 skills (`goal-quench` Phase-3 Stop hook, `hub-cc-pr-reviewer` CC session context, `install-wizard` settings.json write) require Claude-Code-native runtime and are **methodology reference only** under Codex unless a dedicated adapter exists. Use Codex's native goal/session features for goal control, and use `fh-gate` after completion for FH quality gating.
+M3 skills (`goal-quench` Phase-3 Stop hook, `harness-pr-reviewer` CC session context, `install-wizard` settings.json write) require Claude-Code-native runtime and are **methodology reference only** under Codex unless a dedicated adapter exists. Use Codex's native goal/session features for goal control, and use `fh-gate` after completion for FH quality gating.
 
 ### 4. No token accounting
 Codex token usage is billed in the Codex CLI quota and is **not** recorded in any FH session log or orchestrator measurement. Cross-family runs (Gemini/Codex) are invisible to FH's token-budget tooling by construction.

package/knowledge/shared/harness-core/agents_md_runtime_details.md

@@ -178,7 +178,7 @@ is not the headless substitute.
 |---|---|---|
 | **M1 — Full** | No Claude-native dependency | `token-budget-gate`, `asset-placement-gate`, `phantom-quench`, `deep-clarify`, `convergence-loop` |
 | **M2 — Partial** | Core works; native agent/slash-command steps need adaptation | `deliberation`, `steel-quench`, `harness-doctor`, `context-doctor`, `sim-conductor`, `harvest-loop` |
-| **M3 — Claude-only** | Requires a Claude hook or session-scoped dispatch | `goal-quench`, `hub-cc-pr-reviewer`, `install-wizard` |
+| **M3 — Claude-only** | Requires a Claude hook or session-scoped dispatch | `goal-quench`, `harness-pr-reviewer`, `install-wizard` |
 
 **Which phase needs adapting** — the operative half of the M2/M3 rows. Without this a tier label
 tells a non-Claude runtime that a skill is "partial" but not *where* to intervene, which is the only
@@ -191,7 +191,7 @@ move, these per-skill cues did not, and they existed in no other file.)
 | `harvest-loop` | the git-scan phase | PR auto-proposal |
 | `deliberation` | proposal/synthesis structure | Mediator and Jury agent steps |
 | `goal-quench` (M3) | — | Phase 3 depends on a Claude Stop hook |
-| `hub-cc-pr-reviewer` (M3) | — | needs Claude session context |
+| `harness-pr-reviewer` (M3) | — | needs Claude session context |
 | `install-wizard` (M3) | — | writes `settings.json` |
 
 For M2, replace `Agent(subagent_type=...)` and slash-command steps with `fh-run` or direct

package/knowledge/shared/harness-core/claude_md_gate_details.md

@@ -173,6 +173,32 @@ destroys live state without anyone noticing. This is why the loss class is calle
 
 ---
 
+### Scholarly deposit (Zenodo / DOI / arXiv) — measured 2026-09-07, why Step 1b exists
+
+Two things happened on the same day, on the same record (`10.5281/zenodo.22542168`, v1.0.1):
+
+1. **Form ≠ server.** The rich-text description and the companion-DOI related identifier were visible
+   in the deposit form and **absent** from `/api/records/<id>/draft`. The editor had not flushed its
+   state to the server. Nothing in the Pre-Publish gate covered this surface; a hand API read caught it
+   minutes before Publish.
+2. **The machine fields outlive the PDF.** v1.0.1 is a *corrective* release: its body fixes eleven
+   misattributed references. Its Zenodo `references` field still carried **all eleven** pre-correction
+   attributions — the exact strings the release existed to retract — because the PDF was replaced and
+   the metadata was not. `references` / `related identifiers` are what DataCite and citation graphs
+   consume; the PDF is what a human opens. Fixed by editing the record (22 → 24 entries, verified
+   server-side, DOI unchanged).
+
+Consequences that became the four Step 1b items: read the draft through the **service's** API (Zenodo
+InvenioRDM `/api/records/<id>/draft`, legacy `/api/deposit/depositions/<id>`, figshare
+`/v2/account/articles/<id>`), compare against the text you pasted (string vs JSON), md5 the file, and
+on a corrective release diff the machine fields too. The post-publish read is a **detector**, not a
+gate — a wrong field there is fixed by a new corrective version, never silently.
+
+Salience check (same day, floor tier, blind, reps 3, one variable — the edited text injected into a
+clean clone via `--setup`): before 0–1/3 → after 3/3 on all four items. ⚠️ The first sim run was void:
+`sim_isolated_run.sh` clones **HEAD**, so uncommitted edits were absent from every arm — it measured
+the pre-change tree. Recorded so the next author injects the working tree instead of trusting the clone.
+
 ## §Pre-Publish-Hook-Coverage
 
 **Hook coverage — three distinct actions** (refined 2026-06-17 for (a)/(b); (c) added 2026-06-27):

package/knowledge/shared/harness-core/fh_three_layer_canon.md

@@ -566,7 +566,7 @@ standpoint:            ⓑ 는 **자기 필드**가 정본이고, `axes-run` 은
 | **질문하기** | `external-grounding` | ④ 프런티어→조직 전파 |
 
 - **영혼 = 정체성 선언이 아니라 판단의 좌표계**다(무엇이 성공 · 어디로 기움 · 범위 밖 ·
-  절대 안 함). 페르소나 선언은 105런 실측에서 **약모델 순손실**이므로 넣지 않는다.
+  절대 안 함). 페르소나 선언은 **v2a 팔(12런)** 실측에서 **약모델 순손실**이므로 넣지 않는다(🟥 이 발견은 2×2 팩토리얼 81런이 아니라 `CORE IDENTITY` 절만 제거한 **12런 단일변수 팔**의 것이다 — 「105런 실측」으로 부르면 근거를 9배 부풀린다).
   형식 검사기: `scripts/judgment_circuit_lint.sh`.
 - **영혼은 한 번에 안 만들어진다** — FH 는 **씨앗 초안**만 주고, 실제로 채워지는 것은
   현장 하네스가 사용자에 의해 무수히 사용되면서다. 등급 대응: