ForgePlan-marketplace

ForgePlan Hindsight memory plugin for Claude Code. Long-term cross-session memory wrapping Hindsight (vectorize-io). Provides 13 MCP tools (memory_* + mental_model_* + document_ingest_*), 3 auto-hooks (recall on UserPromptSubmit, retain on Stop, force-retain on SessionEnd), and 5 helper skills (/status, /bootstrap, /mental-model, /diagnose, /export-bank). Three activation modes: plugin install (default-on, bank derived from cwd), per-project setup CLI (explicit `.mcp.json`), or direct MCP wire-up (standalone bundle). Source code colocated under plugins/fpl-hsmem/src/. Implements Ruflo-style outcome-feedback patterns (NOTE-004).

Flagship workflow plugin for the ForgePlan ecosystem. 40 engineering skills + dev-advisor agent + 5 hooks (+ smith greeting + insight watchdog). v1.49.0: NEW /forge-heal skill — PRD-074/RFC-020 brick 2 (Layer 4 heal-or-task): propose-not-silent 3-tier dispatch of findings onto PRD-032 AUTO/ADI/USER — AUTO auto-fixes ONLY 2 reversible kinds (phase_mismatch, complete stuck_draft) after a batch confirm + journal + one-call rollback (phase rollback via phase_advance --to prior, NOT undo_last); ADI reasons; USER writes a tracked NOTE-013 task + suggests the agent; never silent-mutates the tree. v1.48.0: NEW /forge-insight skill + insight-watchdog SessionStart hook — PRD-074/RFC-019 brick 1 self-aware integrity: filtered anomaly digest (silent when clean), accurate session counts from forgeplan truth (retires the stale-grep draft-count falsehood), NOTE-013 machine-findings ledger; plus a decay-reminder.sh stale --output/--kind flag fix (its ADR-trigger scan was silently no-opping). v1.46.0: NEW /conformance-vectors skill (SDD optional enrichment — ADR-008 / PRD-072 FR-006 / DEFER-012): the `## Conformance Vectors` convention (frozen language-neutral corpus = sole oracle) + a corpus-agnostic generic harness (freeze + conformance + cross-language equivalence gates, live negative-control-verified) + the proven 4-language semver reference promoted from sandbox (EVID-119) + a cross-language equivalence guardian-gate for multi-language SPECs. v1.45.0: /autorun RIPER methodology gate (DEFER-016 / RFC-018 FR-4) — a Row-4 (non-trivial production bug → RIPER, hook-gate=No) task makes the Plan→Execute transition human_required; autopilot HOLDS before the first Execute dispatch (unattended default = WAIT, not proceed) + checkpoint + CONCERNS EVIDENCE; closes the EVID-167 F1 / G8 autonomous-RIPER gap. v1.44.1: /riper FR-5 pin-hash basis clarified — hash the Research NOTE body content (sections, excluding mutable frontmatter), NOT the projection file (which false-stales on unrelated activation/link/score); + F2 note (the orchestrator records the Research NOTE; research-analyst is read-only Profile C) — DEFER-018 / EVID-172 F1+F2. v1.44.0: RIPER instance #4 (RFC-018 / ADR-010) — /riper formalized into the contract instance (C1-C6 + non-freezable Research C4+C6 + Plan-gate pin-freshness + per-mode hard constraints + hook-gate=No main-session orchestration), smith Row 4 reframe + disambiguator, /methodology-check Step 10 recognizes the Research ## Pinned revision C6; the ADR-012 internal codename is rendered hook-gate in all shipped artifacts. v1.43.0: smith routing-map Row 3 (new feature in existing service) reframed as the explicit SPARC instance — master-coordinated sparc-orchestrator dispatch + mandatory per-gate independent C4 + Pseudocode-absorption + Refinement→TDD delegation + SPARC/BMAD/TDD/Strangler disambiguator (RFC-016/ADR-010 instance #3). v1.42.0: smith routing-map Row 1 (greenfield) reframed as the explicit BMAD instance — master-coordinated agents-bmad dispatch + 3 mandatory C4 gates + BMAD/SPARC/TDD/Strangler disambiguator (RFC-013/ADR-010 instance #2). v1.41.1: smith routing-map adds a 'TDD-first feature' context row routing to the agents-tdd dispatch sequence (RFC-012/ADR-010). v1.41.0: /spec-author skill (ADR-008 light-path SDD; frozen #### Scenario = oracle) merged alongside the FR-3 ground-truth clause (PRs #130 + #131). v1.40.0: RFC-011 FR-3 ground-truth verification clause (generator≠verifier) — AGENT-AUTHORING-GUIDE adds canonical 'Profile B Step 4.5' section (variants A/A'/A'' + `## Ground-truth verification` EVID template + Profile B universal HARD RULE 8 making ML-13 enforceable: empty git diff on a claimed change = BLOCKER even when tests are green); smith routing-map adds 'Ground-truth verification (generator≠verifier)' methodology card applying to every Build/Audit row; corrected the worktree under-claim (isolation works for standalone subagents + Workflow + AgentTeams teammates, not coder-only — real risks are leak + assume-without-verify). Refs PROB-002/RFC-011/ADR-009; smoke `sandbox-verify/r3-reviewer-groundtruth-smoke.sh`. v1.38.1: cookbook §07 fix — final-e2e found CLI vs MCP agent-id validator asymmetry on forgeplan 0.32.1 (filed as forgeplan#353). Cookbook §07.1 example switched from 'adr-architect/1.11.1' (CLI rejects) to dash-form 'adr-architect-v1-11-1' (both accept). §14.4 added documenting the asymmetry + workaround. §14.5/14.6/14.7 renumbered. v1.38.0: NEW `/forgeplan-cookbook` skill — agentic-RAG router for 66 mcp__forgeplan__* MCP tools + 82 forgeplan CLI subcommands. 14 sections by use-case (lifecycle / discovery / quality / pipeline-health / session-phase / AI-driven / multi-agent / brownfield / activity / FPF / playbooks / plugins / release / MCP-safety) + flat alphabetical _index. Quick router maps intent → section. Section 14 documents 3 filed upstream bugs (#348 score-all hint, #350 CRITICAL @filepath data loss, #351 plugins doctor syntax) + canonical safe patterns. v1.37.1: CRITICAL safety convention — AGENT-AUTHORING-GUIDE now documents the MCP `body` parameter as literal-string-only (forgeplan#350 silent data-loss bug). Canonical safe pattern: Read file via host CLI, pass content as string. CLI vs MCP semantics table prevents agent confusion. Profile A/B/D agents are most exposed (canonical body writers). v1.37.0: --help-driven verification + Step 1b uses native `forgeplan migrate-secrets --apply` for LLM key import (PRD-077 FR-023 Part B). 4-row table covers: gemini (default template match), anthropic, openai, or no-key-found path. Removes 35-line manual config snippet — auto-import is the canonical forgeplan flow. v1.36.0: deep-e2e fix pack — (a) templates/smith-bootstrap.md acceptance probes use --type (4 places) — was --kind=, drifted from SKILL.md fix in v1.35.2; (b) new Step 1b in smith-bootstrap probes .forgeplan/config.yaml for an llm: block right after forgeplan init; if absent, prints informational guidance (3 providers — Anthropic / Gemini / OpenAI — with full config + secrets.env example) so the user can configure before the first /forge-cycle Step 4.5 ADI call; (c) Failure modes covers downstream LLM-not-configured halt. Closes deep-e2e findings 2 and 4. v1.35.2: e2e-test fix pack — smith-bootstrap Step 5/6 acceptance probes use correct CLI flag `--type` (not `--kind` — that name only exists in MCP forgeplan_new(kind=...) form), Step 6 documents the literal section names forgeplan validate MUST-requires (Problem statement, Target audience, Goals/Success Criteria, Functional Requirements with literal capitalisation, Out of scope, Related Artifacts) + cites the validator rule names that trigger on miss, and adds `forgeplan validate PRD-NNN → PASS` to acceptance criteria. v1.35.1: smoke-test fix — Step 0c 6th state `plugin-handles-it` (fpl-hsmem enabled + no project-local .mcp.json hindsight block = normal/healthy, not unclassified). Smoke against fresh tmp project surfaced that the previous 5-state matrix marked this normal case as 'UNCLASSIFIED'. Reference implementation now ships as bash if/elif ladder with explicit precedence order. v1.35.0: MCP-wiring drift closure trio — (1) /fpl-init Step 5 migrated from Python merge to native `forgeplan mcp install --client claude --scope project` (same primitive as smith-bootstrap Step 0b, eliminates skill-to-skill drift); (2) hooks bundle expanded from 1 to 3 files (safety-hook + pre-pr-evidence-check + forge-safety-hook) at templates/hooks/; (3) AGENTS.md MCP server registration section now leads with `forgeplan mcp install` as canonical wiring command across Claude Code / Cursor / Windsurf / Gemini / Codex. v1.34.0: smith-bootstrap Step 0 refactored from halt-and-instruct to active infrastructure runner. 5 sub-steps: 0a plugin enablement gate (halt-only on MUST miss), 0b forgeplan MCP auto-wire via native `forgeplan mcp install --client claude --scope project` (replaces halt + Python merge), 0c Hindsight detection with 5-state classification (fully wired / key missing / env-var naming mismatch / Docker quick-start / plugin not enabled) — informational guidance, never writes secrets, 0d minimal hooks bundle copy from templates/hooks/ (safety-hook.sh starter), 0e user-driven /reload-plugins + ToolSearch verify. Closes real-project gap observed 2026-05-20 where bootstrap landed CLAUDE.md+AGENTS.md but skipped MCP wiring → downstream agents shell-fallback → degraded artifacts. v1.33.0: smith-bootstrap pre-flight gates. v1.31.1: EPIC-002 niceties patch. v1.31.0: EPIC-002 — 4 smith skills + templates + session-start.sh.

Practical cookbook of forgeplan CLI recipes — task-oriented how-to guides covering getting-started, PRD authoring, Evidence patterns, workflow cycles, multi-plugin discipline, AI-pair dispatch, troubleshooting Sprint A-O anomalies, and CLI/MCP quick-reference. Each recipe: Цель → Команда → Пример → Common errors. / Практическая поваренная книга рецептов forgeplan CLI — пошаговые руководства по онбордингу, PRD, Evidence, рабочим циклам, мультиплагинной дисциплине, диспетчеризации AI-агентов, устранению аномалий и справочнику CLI/MCP.

Agentic RAG methodology for Claude Code skills — how to structure large knowledge bases with SKILL.md as a thin router + sections/_index.md + 30-50 line content files. Use when building a skill with >300 lines or >5 topics. Includes 5+ anti-patterns, copy-pasteable starter-kit, and distribution recipes. Triggers on: agentic RAG, SKILL.md structure, sections pattern, skill methodology, knowledge base design, how to write a skill.

UX laws knowledge base and frontend code reviewer. Analyzes HTML/CSS/JS/React/Vue/Svelte code against 30 Laws of UX principles. Provides actionable recommendations for improving user experience. Use when building UI components, reviewing frontend code, or learning UX best practices. v1.4.1: ux-reviewer agent canonical-lint compliant (LR-1..LR-3).

Structured engineering workflow for forgeplan users. Provides /forge-cycle (full SDLC cycle: Hindsight v2 memory bootstrap + parallel subagent dispatch via forgeplan_dispatch + Task tool + multi-reviewer audit with 4 parallel agents + mail-as-beads autonomous decisions logging + MCP-first per PRD-021/022 + claim/release per PRD-020 — implements RFC-003/PRD-025) and /forge-audit (multi-expert review), methodology knowledge base, quality gate hooks. Requires forgeplan CLI. v1.12.0: Sprint Z7 PRD-059 — /forge-cycle Step 4.5 FPF ADI mandatory for Standard+ (≥3 hypotheses required; guardian BLOCKER when no ADI EVID). v1.11.0: Sprint Z6 PRD-057 — /forge-cycle Step 6.5 BMAD adversarial review mandatory for Standard+ (guardian BLOCKER when zero Profile B EVID). v1.10: PRD-034 Sprint F — /forge-cycle template clarified (Anomaly #1 fix) + Phase 6.7/Step 7.7 live-smoke discipline check (Anomaly #11 wiring). v1.9.1: forge-advisor agent canonical-lint compliant.

[DEPRECATED — superseded by fpl-skills] Universal engineering toolkit: multi-expert audit, wave-based sprints, session context recall, card-based structured reports with auto-trigger hooks, and safety hooks. Works with any project and language. New users: install fpl-skills@ForgePlan-marketplace instead. v1.6.3: dev-advisor agent canonical-lint compliant (LR-1..LR-3).

First Principles Framework (FPF) — thinking amplifier for structured reasoning, system decomposition, and decision-making. Enhanced version with applied patterns, forgeplan integration, and practical commands. Based on the FPF specification by Anatoly Levenchuk. v1.4.1: fpf-advisor agent canonical-lint compliant (LR-1..LR-3).

Unified workflow: Forgeplan artifacts + Orchestra task tracking + Claude Code AI execution. Bidirectional sync, Session Start Protocol with Inbox Pattern, and methodology knowledge base. Requires forgeplan CLI (private app, access via project admin) + Orchestra MCP server. v1.4.1: orchestra-advisor agent canonical-lint compliant (LR-1..LR-3).

Brownfield extraction pack for forgeplan — turns legacy code + docs into a structured forgeplan graph. Ships 12 extraction skills (ubiquitous-language, use-case-miner, intent-inferrer, invariant-detector, causal-linker, hypothesis-triangulator, interview-packager, scenario-writer, kg-curator, canonical-reproducer, reproducibility-validator, rag-packager), 2 orchestration playbooks (extract-business-logic, phase-transitions), 3 integration recipes (autoresearch hooks, forgeplan MCP additions, RAG export), 5 mappings (c4-to-forge, ddd-to-forge, madr-to-forge, obsidian-to-forge, autoresearch-to-forge), templates, examples, and methodology docs. Implements two-tier extraction (Factum vs Intent) with confidence taxonomy and ADI cycle. v1.4.0: Sprint V (PRD-048) — discover agent migrated from standalone to plugin (canonical Profile A pattern, MCP-first 7-phase procedure, 9 brownfield MCP primitives wired).

v1.4.1: coder body gains a 'TDD GREEN-phase discipline' section (RFC-012 FR-4) — when dispatched in a TDD GREEN phase, never Write/Edit test files; on a wrong test STOP and emit TEST_BUG; lint after each change. v1.4.0: code-reviewer + tester (Profile B) bodies gain Step 4.5 Ground-truth verification clause (RFC-011 FR-3) — reviewers MUST read frozen git diff themselves before PASS; empty diff on a claimed change = BLOCKER (vacuous green). Core development agents: debugger, code reviewer, error detective, performance engineer, production validator, plus a complete dev team (coder, planner, researcher, reviewer, tester, TDD London School). v1.3.1: code-reviewer + tester (Profile B) bodies patched with Step 9b NEEDS_ACTIVATION sentinel emit instruction per Sprint E PRD-033 — Profile B reviewers now emit `<<NEEDS_ACTIVATION: EVID-XXX>>` on first line of return to enable orchestrator auto-activation. v1.3: 8 legacy specialists canonical-lint compliant. v1.2: coder Profile C-coder + code-reviewer + tester (batch 2 forgeplan-aware migration).

Domain-specific language and framework specialist agents. Provides expert-level guidance for TypeScript, Go, React, Next.js, Electron, embedded systems, fullstack, game dev, mobile, and WebSocket engineering. v1.1: 11 agents canonical-lint compliant (LR-1 model:sonnet, LR-2 hex colors, LR-3 bilingual descriptions with triggers).

Professional specialist agents: security experts, architecture reviewers, creative tools, research analysts, infrastructure engineers, plus pipeline-canonical roles. v1.12.0: RFC-011 FR-3 — ground-truth verification clause (Step 4.5 + HARD RULE + ## Ground-truth verification EVID section) added to 5 Profile B reviewers (security-expert, architect-reviewer, system-dev, evidence-recorder, artifact-reviewer) + guardian gate row; empty git diff on a claimed change is a BLOCKER even when tests are green (PROB-002 / ADR-009 / ML-13). v1.11.0: EPIC-003 / Sprint AA PRD-067/069 — guardian Step 5 verdict matrix +2 new rows: C4 CONCERNS (≥3-module ADR without docs/c4 file, Sprint Z9 PRD-060 implementation) + OpenSpec BLOCKER (supersede without ## Delta-spec, Sprint Z8 PRD-058 activation gate). Closes G2+G4 methodology auto-enforcement gaps. v1.10.1: EPIC-002 niceties patch. v1.10.0: EPIC-002 PRD-062 — new `smith` master-orchestrator agent (Profile B-orchestrator). v1.9.4: Sprint Z9 — adr-architect C4 dispatch. v1.9.3: Sprint Z7 — guardian ADI BLOCKER row.

GitHub operations agents: PR management, issue tracking, release automation, multi-repo coordination, project boards, workflow engineering, and repo architecture. v1.1: 7 agents canonical-lint compliant (LR-1..LR-3) — sonnet model, hex colors, bilingual EN/RU/Triggers descriptions.

v1.3.2: ADR-012 internal codename rendered hook-gate (was a Russian codename) in shipped artifacts — sparc-orchestrator + /sparc skill; terminology only, no behavior change. v1.3.1: /sparc skill Step 2 reframed - hook-gate=No means the MAIN SESSION orchestrates the walk (dispatches phase agents directly); a dispatched sparc-orchestrator subagent cannot nest-dispatch (it is a planner/guardrail, not an executor) - per the RFC-016 live dogfood (EVID-165). SPARC development methodology agents: orchestrator with quality gates, plus 4 phase specialists (specification, pseudocode, architecture, refinement/TDD). v1.3.0: THIRD instance of the AD/AID-PDLC sub-cycle contract (ADR-010 / RFC-016) — sparc-orchestrator finished to a contract-conformant Profile B-orchestrator (disallowedTools denylist + Task-dispatch + C1-C6 framing; closes the LR-8 tools allowlist / EVID-153 F2) + NEW /sparc skill (5-phase master-coordinated walk: Specification→Pseudocode→Architecture→Refinement→Completion, mandatory per-gate independent C4, Pseudocode-absorption, Refinement→TDD delegation). hook-gate=No — no hook, no lib, no init. v1.2.2: refinement notes enforced-TDD delegation (RFC-012) — RED test-writing delegates to agents-tdd:coder-tdd and GREEN to agents-core:coder; refinement retains refactor-after-green. v1.2: ALL 5 agents canonical-lint compliant — sparc-orchestrator + pseudocode + refinement migrated to model=opus|sonnet, hex colors, bilingual EN/RU/Triggers descriptions (legacy lint warnings eliminated). 100% of agents-sparc pack passes LR-1..LR-3. specification + architecture remain forgeplan-aware Profile A (PRD-026 Phase 3 batch 1).

v0.2.1: tdd-orchestrator denylist adds forgeplan_validate - B-orchestrator parity (DEFER-015 / EVID-157 F1): a master coordinates, it does not validate; now matches smith/sparc + the AGENT-AUTHORING-GUIDE canon. Enforced-TDD methodology: tdd-orchestrator master + RED/GREEN phase agents + fail-closed PreToolUse gate (first instance of the AD/AID-PDLC sub-cycle contract). Implements RFC-012 / ADR-010: tdd-planner (scenarios→plan) + coder-tdd (plan→failing tests, RED) + tdd-test-validator (independent C4 verifier) + PreToolUse tdd-gate (permissionDecision:deny, exit2 fail-closed) + normalized full-file SPEC hash freeze (FR-6).

v0.1.2: ADR-012 internal codename rendered hook-gate in shipped artifacts — bmad-orchestrator + bmad-gate.sh comment; terminology only, no behavior change. v0.1.1: bmad-orchestrator denylist adds forgeplan_validate - B-orchestrator parity (DEFER-015 / EVID-157 F1); now matches smith/sparc + the AGENT-AUTHORING-GUIDE canon. BMAD greenfield methodology: bmad-orchestrator master walks the persona arc (Analyst → PM → Architect → Scrum-Master → Dev → QA) with a blocking quality-gate at every handoff and a fail-closed no-code-before-plan PreToolUse gate (second instance of the AD/AID-PDLC sub-cycle contract). Implements RFC-013 / ADR-010: reuses existing forgeplan-aware personas (brief-intake/specification/adr-architect/architecture/goal-planner/coder/tester/code-reviewer/guardian/evidence-recorder) + a new bmad-orchestrator + /bmad + /bmad-init + bmad-gate (permissionDecision:deny, exit2 fail-closed) reading a per-branch phase state file.

Claude Code ecosystem reference — opinionated agentic-RAG guide for CLAUDE.md, plugins, agents, hooks, MCP, and anti-patterns. Synthesises 47+ audit findings from ForgePlan marketplace work. v1.1.0: all 6 sections authored — claude-md + plugins + agents + hooks + mcp + anti-patterns (RFC-005..009 closed / DEFER-005..009).

v0.19.0: forgeplan-web dogfood re-run follow-ups (findings A/B/C, after v0.18.0 proved the pipeline completes end-to-end on a real repo -- top map confirmed + 4/4 layers confirmed, GC-9 + parent_zone live-PASS). (A) map-emit.mjs input PRE-FLIGHT: refuse a wrong-shape (pre-v0.18) scratch instead of silently degrading. Resuming an old scratch under the new script had nested repo_head/project under meta. + a flat {edges:[]} shape, so the emit silently produced a 'nogit:' fingerprint, dropped title/description_ru, and would have shipped 0 edges -- against the pack's honest-failure discipline. The script now names the mismatch (repo_head/project must be top-level; edges must be {typedLink,codeDep} with id+namespace) and writes NOTHING; +2 test/emit-script.mjs scenarios. (B+C) scoped-layer ROLE DECOMPOSITION now mandatory in the orchestrator Scoped-pass rules + /map-build-layer + a map-emitter HARD RULE: a layer build routes input-prep through zone-extractor (scoped .extract.json + persisted .composition.json) + edge-verifier (scoped .edges.json), and map-emitter does ONLY flows + the script -- it never self-authors scoped inputs (the dogfood saw 3/5 emitters overstep, and one self-authored layer forgot repo_head -> a nogit: fingerprint on that one layer). 81 assertions green across 4 suites. Plugin v0.18.0 -> v0.19.0, catalog 1.117.0 -> 1.118.0. v0.18.0: DETERMINISTIC EMIT -- the headline fix from the 2026-07-15 forgeplan-web dogfood, which proved the LLM-typed emitter cannot scale to a real repo. On 274 nodes / 316 edges (~4,000 lines) map-emitter blew the 64,000-token output cap 3 runs out of 3; round 3 suppressed prose to five lines and STILL died, proving THE DOCUMENT is the cap-breaker, not agent chattiness (round 1 only 'barely fit': 553k tokens, 56 min -- so even the first write was at the edge, and anything larger fails outright). Neither proposed workaround was enough: granting Edit only fixes CORRECTION loops (the initial write stays at the cap), and chunked writes need Edit/Bash anyway while still having an LLM re-type 4,000 lines. Root cause is architectural -- an LLM re-typing a mechanical document whose every content DECISION was already made upstream (.extract.json=zones/nodes, .edges.json=edges, composed composition=layout). NEW scripts/map-emit.mjs (deterministic, dependency-free Node, mirroring map-guardian.mjs) now owns everything clerical: meta + 'git:<sha>' build anchor, tier-row layout + CM-22 accent de-collision, CM-07 canonical layer meta + B5 CONTENT seed_fingerprint, CM-01 visible() collapsed-child rewrite + CM-05 edge_ids resolution, CM-23 _content_sig drop + found_at hoist, the assembly-guard trio with reject-before-write, atomic tmp-rename, status:'proposed', and the NEEDS_CONFIRM sentinel. The map-emitter AGENT now decides ONLY the flows (the sole judgment call) into a tiny .work/.emit-plan.json and invokes the script (Bash granted for that one purpose; Edit + every forgeplan_* mutator still denied; it no longer writes map.json at all). The script's plain fs write is invisible to the write-hook BY CONSTRUCTION -- the same sanctioned shape ADR-017 blesses for the guardian -- and GC-5's git audit remains the single-writer backstop. Orchestrator SELECT now persists the composed base+overlays as .work/.composition.json (the script parses no YAML). NEW test/emit-script.mjs (35 assertions) including the end-to-end one that matters: emit a document, then run the REAL guardian over it and require exit 0 (GC-2/GC-7/GC-10/GC-11 PASS, GC-9 PASS on a layer) -- proving emitter and gates agree. No token cap, deterministic, instant, scales to any repo. Plugin v0.17.0 -> v0.18.0, catalog 1.116.0 -> 1.117.0. v0.17.0: first-live-dogfood fix -- /map-doctor now handles a LEGACY (pre-v0.16) map gracefully. Running the v0.16.0 /map-doctor on a REAL forgeplan-web map (generated by the OLD v0.7.1 map-pack; source_fingerprint 'fpw-map-...' is not a git anchor) surfaced 2 real command-prose defects the deterministic gates + the self-audit could NOT catch -- they test v0.16-generated maps, not legacy ones. (A) A legacy map produced a wall of GC-7/GC-9/GC-10 BLOCKERs that READ as corruption but are just schema drift (the old generator never emitted found_at / canonical layer-meta / grouped-mega-kind); /map-doctor now DETECTS a legacy map (non-git source_fingerprint, or layers lacking scope:'layer'/parent_map_id) and LEADS with a 'LEGACY MAP -> rebuild once with /map-build' banner, tagging those BLOCKERs legacy-schema-gap instead of a dozen alarming per-layer failures. (B) A layer with a MISSING meta.parent_zone was wrongly flagged an ORPHAN; orphan now means ONLY a present-but-dangling parent_zone (a zone deleted from the top map), while a missing parent_zone is non-canonical meta (GC-9's job), not orphan. The doctor itself verified WORKING on the live run (fast, read-only, accurate 11+2 checks, readable digest, correct fix commands) -- the real-run gap is now closed for /map-doctor. Plugin v0.16.0 -> v0.17.0, catalog 1.115.0 -> 1.116.0. v0.16.0: dogfood-brief batch 5 (CM-25) -- two new commands + content-aware staleness, closing the 25-finding v0.7.1 dogfood brief. Folds in an 18-finding adversarial self-audit remediation (a generator!=verifier workflow over the 5 batches of agent-prose): reconciled the E5 auto-cascade seed_fingerprint to the content-aware B5 form (it still read as membership-only, re-introducing the exact staleness-blindness B5 fixed on the default --layers path), gave the guardian a `--check-only` read-only deep flag so /map-doctor --deep exercises Layer B WITHOUT the proposed->confirmed write it previously performed (+ guardian-layer-b.mjs Scenario 7), realigned the zone-extractor AGENT body to the grouped-collapse contract its own SKILL uses (the body still taught the 170->1 single-blob dump), dropped the unfulfillable 'fold the narrating doc blob SHA' staleness claim (docs-scanner has no Bash to compute one), threaded manifests[] to edge-verifier, fixed the _content_sig/contentSig producer/consumer key drift, and corrected ~9 cross-file prose drifts (smoke-coverage GC-7..GC-11, stale GC enumerations, mtimes vs git-anchor, mega ref prefix, project shape). NEW /map-refresh: rebuild ONLY the layers whose content changed since the map was built, not the whole cascade. It rests on a real upgrade -- meta.seed_fingerprint is now a CONTENT fingerprint (sha1 over each member's id + its git blob signature / artifact updated-stamp), not the old membership-only id-set hash; so EDITING a file inside a card (git) or updating a doc now marks its layer stale, which the v0.8.0 idempotent-skip was blind to. The top map's meta.source_fingerprint becomes a robust git build anchor 'git:<HEAD-sha>' (was fragile mtimes); /map-refresh diffs `git diff <anchor>..HEAD` UNION uncommitted, maps changed files to stale zones via node provenance.ref paths + doc narration, recomputes each layer's content fingerprint, and rebuilds only the mismatched layers in parallel (the orchestrator's new Refresh mode). code-scanner records per-file git blob content_sig + repo HEAD; forgeplan-scanner records per-artifact updated; zone-extractor threads a scratch-only _content_sig (dropped before ship, CM-23) that map-emitter folds into seed_fingerprint. NEW /map-doctor: fast, read-only triage -- runs the deterministic guardian (--smoke Layer A) over the top map + every layer, flags stale layers via the same git anchor, finds orphan layers (zone gone), and prints a fix-oriented digest (which command repairs each issue) WITHOUT regenerating anything or dispatching an agent. Doctor diagnoses, refresh fixes. All agent-prose + 2 command files; no guardian/schema change. Plugin v0.15.0 -> v0.16.0, catalog 1.114.0 -> 1.115.0. v0.15.0: dogfood-brief batch 4B (CM-10/CM-15/CM-21/CM-22/CM-23/CM-24) -- layout + grouping prose, completing batch 4 (the 25-finding v0.7.1 dogfood brief is now closed except batch 5 = new /map-doctor + /map-refresh commands). CM-24: edge-verifier dedupes edges globally by (from,to,relation) (the edge id's own key), keeping the higher-trust survivor -- a relationship the graph AND a grep both surfaced no longer ships twice or double-lights a flow (A->B and B->A stay distinct). CM-22: map-emitter tier-row layout de-collides accents -- no two grid-neighbour zones share a --map-accent-* token (GC-11); minimal reassignment keeps authored accents and only breaks a colliding tie. CM-15: zone-extractor collapses at EVEN density -- ~ceil(sqrt(N)) comparably-sized group cards (sub-split a lopsided giant), never one blob next to slivers. CM-10: the decision zone can group by narrative ARC (Shape/Design/Decide/Prove) on the TOP map when the composition provides decision_grouping:"arc" + kind_to_arc (promoting the arc derivation the extractor already did one level down), instead of a bag-of-kinds. CM-21: a scoped layer build selects its sub-composition from a catalogue keyed on the target zone's KIND (store->data-shaped, surface->UI-shaped, truth->decision-arc) instead of flooring every drill-down to generic. CM-23: field discipline -- emit only schema-defined + named-consumer fields uniformly (all-or-none); drop speculative extras like the inconsistent description_ru_source / edge note the dogfood shipped. All agent-prose; GC-7..GC-11 gates + 3 test suites catch regressions. Plugin v0.14.0 -> v0.15.0, catalog 1.113.0 -> 1.114.0. v0.14.0: dogfood-brief batch 4A (CM-11/CM-16/CM-17/CM-19/CM-20) -- flows + labels prose. CM-19: a flow is a CONNECTED edge-path, not an arbitrary node set -- map-emitter walks the edge graph and, when a hint's path splits into two edge-disconnected arcs, emits two separate flows instead of one spliced one (the v0.7.1 decision flow lit two islands with a dead gap). CM-11: every flow (hint- and entrypoint-derived) carries a short EN chip name + RU steps (GC-8). CM-17 + CM-20: ONE label policy (zone-extractor Algorithm 5c) -- a code node's label is the basename/curated short name (full repo-relative path in provenance.ref), an artifact node's label is "<ID> -- <title>" (bare artifact_id in provenance.ref); the full machine ref never renders on the card. CM-17: forgeplan-scanner now carries title + a one-line summary per artifact; zone-extractor builds the artifact node's description_ru from that summary (omit when absent, never faked). CM-16: a scoped layer build narrates at FULL budget (one zone = room to explain each node richly) and the richer description propagates up so the top card never carries a thinner description than its own layer produced. All agent-prose; the GC-7..GC-11 gates + 3 test suites catch regressions. Plugin v0.13.0 -> v0.14.0, catalog 1.112.0 -> 1.113.0. v0.13.0: dogfood-brief batch 3 (CM-06/CM-07/CM-08/CM-18) -- content-provenance + meta canonicalization prose the v0.11.0 GC-7..GC-11 gates guard against regression. CM-06: real `found_at` -- code-scanner gains a NARROW git-read Bash (git log --diff-filter=A first-add date, argv-safe, read-only; same edge-verifier precedent + GC-5 git-audit backstop, user-approved) to record per-file first_seen; forgeplan-scanner records each artifact's `created`; zone-extractor Algorithm 5b stamps found_at from a REAL first-seen (code node=git first_seen, artifact node=created, mega=earliest child, deterministic repo-level fallback) -- NEVER now(), which churns the GC-7 append-stability sort key (the v0.7.1 constant/missing-found_at defect). CM-07: map-emitter Algorithm 4 freezes the canonical layer-meta struct GC-9 requires (scope:"layer" + parent_map_id + parent_zone + map_id==<parent>::<zone> frozen :: separator + no needs_confirm floor + seed_fingerprint; the top map writes NONE of these layer keys). CM-08: docs-scanner extracts a project {title, description_ru} from the real README H1 + tagline (omit if absent, never invented); zone-extractor carries it through .extract.json; map-emitter stamps meta.title + meta.description_ru (additive, meta.additionalProperties:true) -- the map finally carries a human heading instead of a bare map_id. CM-18: zone-extractor Algorithm 5 -- a group mega's kind is ALWAYS the literal "mega" (GC-10 BLOCKERs is_mega!==(kind==mega); never the members' leaf kind, which inflated counts) and its provenance.ref carries the machine-recoverable preimage mega:<zoneId>:<groupKey>, not a human sentence. The one security-surface change (code-scanner narrow git-read Bash) is user-approved and lint-safe (code-scanner stays C-coder-class: denies every forgeplan_* mutator + Edit/NotebookEdit). Plugin v0.12.0 -> v0.13.0, catalog 1.111.0 -> 1.112.0. v0.12.0: dogfood-brief batch 2 (P0 prose fixes CM-01/CM-02/CM-05) -- the flow + id-stability defects the v0.11.0 deterministic gates now guard against regression. CM-02: node `kind` is frozen as an ALTITUDE-INVARIANT identity property (zone-extractor skill Algorithm 2a) -- only a node's `zone` is re-computed per scope, so `sha1(kind:ref)` is byte-identical on the top map and inside a drilled-in layer; keeps GC-6 valid and lets forgeplan-web's deriveSubDocument carry a node across altitudes (id-carry no longer breaks when a scoped pass would classify the same entity differently). CM-05: edge-verifier now mints a stable content-hash edge id (sha1("edge:"+from+"|"+to+"|"+relation)[:12] over the REMAPPED endpoints, doubling as the CM-24 dedup key) so flows can reference edges; map-emitter carries edge.id through verbatim and resolves each consecutive flow node pair into flow.edge_ids -- GC-8 had WARNed on the always-empty edge_ids chip. CM-01: after grouped mega-collapse, map-emitter rewrites every flow's node_ids through a visible() map (a hidden collapsed-child -> its containing mega) with consecutive-dedup, so selecting a flow lights RENDERED mega cards the user can expand instead of hidden children (which lit nothing). All three are agent-prose (skills/zone-extractor, skills/edge-verifier, skills/map-emitter + a map-orchestrator scoped-layer note) -- no gate or schema change; the GC-7..GC-11 gates + the 3 test suites catch any regression. v0.11.0: five new deterministic guardian gates (CM-13, from the 5-audit v0.7.1 dogfood brief) -- GC-7 found_at completeness (BLOCKER), GC-8 flow completeness (WARN: multi-node flows must carry edge_ids + RU steps), GC-9 layer-meta canonicalization (BLOCKER: frozen keys + <parent>::<zone> map_id + no auto-confirmed needs_confirm floor), GC-10 is_mega<=>kind==mega (BLOCKER), GC-11 accent-neighbour collision (WARN). All Layer A (doc-only, run under --smoke). They deterministically catch the content defects the dogfood shipped (CM-05/06/07/18/22). Covered by test/guardian-content-gates.mjs (13 assertions). v0.10.0: composition library COMPLETE -- 7 more abstract base archetypes (data-pipeline / ml-project / ai-agent / cms / mobile-app / desktop-app / docs-site), for a full set of 13 base archetypes + 6 pattern overlays. Any common project type -- frontend, backend, CLI, library, data, ML, agentic, CMS, mobile, desktop, docs, fullstack, or the generic floor -- now has a tuned archetype, and hybrids compose base + overlays into one map. v0.9.0: composition BREADTH -- the base-TYPE + pattern-OVERLAY engine (COMPOSITIONS-GUIDE). The orchestrator SELECT stage now detects ALL active pattern overlays independently (not just one base) and composes base + overlays into one effective composition; the emitter lays the merged zones out by tier row (entry/core/data/decisions) for any zone count. Ships 6 overlays (microservices / serverless / cqrs-eventsourcing / mcp-server / api-gateway / monorepo -- the last a regroup-by-package modifier) + 4 new abstract base archetypes (web-frontend / backend-api / cli-tool / library-sdk) alongside web-fullstack / generic (rust-cli-mcp kept as a legacy forgeplan-core specialization). A hybrid repo (e.g. monorepo + backend-api + microservices) now composes into one coherent map instead of falling to the generic floor. v0.8.0: E5 auto-cascade layer generation (PRD-076 FR-5 / ADR-018) -- after the top map is CONFIRMED, /map-build (default --layers) auto-fans-out one scoped per-zone layer build IN PARALLEL (disjoint map/layers/<zone>.json write targets, own scratch + guardian pass), gated by 3 cost controls: seed-fingerprint idempotent skip (unchanged zone not rebuilt), thin-zone threshold (~6 members), depth policy (first level auto; deeper on demand via /map-build-layer or --layers-depth N). meta.seed_fingerprint doubles as forgeplan-web's staleness check. Rejected eager (top map still mutates in gate loops) + sequential (sum-of-zones wall-clock). 8-agent orchestrated pipeline that scans any forgeplan-enabled project and emits a validated .forgeplan/map/map.json (schema forgeplan.map/v1) for forgeplan-web's composed-map renderer. EMITTER profile throughout: a fail-closed PreToolUse hook + agent denylist + a deterministic map-guardian.mjs gate make it structurally impossible for the pipeline to write anywhere except the derived map/layer files or to mutate any forgeplan artifact. v0.7.0: understanding-map quality wave (PRD-076). E1 -- architecture-quality content: over-capacity zones collapse GROUPED by kind (PRD/RFC/ADR/EVID/...) instead of one opaque 170-artifact card; zones carry an abstract RU region description + meaning-subtitle; nodes get a rich RU description synthesized from GROUNDED scanned facts (code-scanner now records module facts -- top-comment, exports, role -- never fabricated from a filename). E2 -- richer flows: web-fullstack ships 4 flow_hints + the emitter derives entrypoint flows to reach 6+ named journeys. E3/E4 -- per-zone generated layers: new /map-build-layer "<zone>" command + orchestrator scoped mode emit a zone's own validated sub-map to .forgeplan/map/layers/<zone>.json (hook + guardian GC-5 extended to sanction map/layers/**); rendering the layer needs forgeplan-web's deriveSubDocument seam (cross-repo contract, PRD-076 FR-3). v0.6.0: output-quality pass on the first rendered real map -- (O-1) node.meta is now a SHORT card subline (full prose -> description_ru), fixing cards that overflowed ~5x; (O-2) the emitter derives flows[] from a composition's flow_hints (short chip name + path + RU steps), so the flow-navigation chips are no longer dark on real data; (O-3) web-fullstack lays zones out in a 2D grid instead of a tall single-column ribbon. Adds COMPOSITIONS-GUIDE.md (the abstract-archetype authoring contract: base TYPE + pattern OVERLAY model, 7-accent convention, shared canvas constants, depth-agnostic hints, flow_hints) and flow_hints on the two curated templates. v0.5.0: dogfood tail-stage fixes -- the write+validate defects the first end-to-end forgeplan-web run surfaced. (F4) map-emitter-gate.sh normalizes the plugin-qualified dispatch identity (forgeplan-map-pack:map-emitter) before its single-writer check -- an exact-string test was denying the emitter's OWN write. (F6) guardian XC-2 + the edge-verifier grep now scope every re-grep with --exclude-dir (node_modules/.git/build/... -- an identical set both sides share) -- an unbounded grep -rlF over the whole repo hung the guardian on any repo with dependencies. (F5) documents honestly that the Write-hook cannot police a Bash-mediated write (edge-verifier needs Bash for grep), so GC-5's after-the-fact git audit is the real single-writer backstop, not the hook. Adds test/emitter-gate-identity.sh + a node_modules-exclusion scenario in test/guardian-layer-b.mjs. v0.4.0: first-dogfood hardening -- fixes the 3 defects the first real forgeplan-web run surfaced. (1) Guardian XC-1 now compares typed-link edges in content-hash id-space (v0.2.0 compared them against artifact-id-keyed scan edges, failing EVERY edge). (2) Guardian GC-5 is now a write-SCOPE check that passes whether map.json is committed (forgeplan-web ships it tracked as the P0 render-proof) or gitignored -- v0.2.0 BLOCKER-ed any committed-map repo and carried a porcelain leading-space parse bug. (3) The web-fullstack composition detection + zone_hints are depth-agnostic, so a nested app root (template/src/**) is recognized instead of falling to the generic floor. Adds test/guardian-layer-b.mjs, a non-smoke regression test covering the Layer-B checks --smoke never exercised (XC-1/GC-5/GC-6/XC-2). v0.3.0: /map-build slash command as the human-facing entry point. v0.2.0: full P1 implementation -- all 8 agents, 3 skills, 3 MVP composition templates, the map-build playbook, and the guardian/hook pair. Pre-1.0: guardian fixes are regression-tested; the depth-agnostic detection fix awaits final confirmation on a full forgeplan-web re-run.