@mmerterden/multi-agent-pipeline 11.1.0 → 11.3.0

package/pipeline/commands/{multi-agent.md → multi-agent/SKILL.md}

@@ -5,13 +5,13 @@ allowed-tools: Agent, Bash, Read, Write, Edit, Glob, Grep, TaskCreate, TaskUpdat
 
 # Multi-Agent Task Orchestrator
 
-> **MUST: Figma MCP is BLOCKING for any task with a Figma reference.** Before any UI synthesis runs in this pipeline (analysis, planning, dev, or rework), call `mcp__claude_ai_Figma__get_design_context` for every referenced frame and use the `CodeConnectSnippet` component name verbatim. Backend-only tasks are exempt. See `refs/rules.md` "User Interaction Discipline" and `pipeline/rules/figma-pipeline.md` "MUST: Figma MCP-first (BLOCKING)".
+> **MUST: Figma MCP is BLOCKING for any task with a Figma reference.** Before any UI synthesis runs in this pipeline (analysis, planning, dev, or rework), call `mcp__claude_ai_Figma__get_design_context` for every referenced frame and use the `CodeConnectSnippet` component name verbatim. Backend-only tasks are exempt. See `$HOME/.claude/multi-agent-refs/rules.md` "User Interaction Discipline" and `pipeline/rules/figma-pipeline.md` "MUST: Figma MCP-first (BLOCKING)".
 
 Parse the user input and route to the correct sub-command.
 
 **Input**: $ARGUMENTS
 
-> **Language (read FIRST)**: Before any status output, read `prefs.global.outputLanguage` from `$HOME/.claude/multi-agent-preferences.json` and render every conversational line in it. `AskUserQuestion` and external payloads stay English. Full contract: `refs/rules.md` "Language Application".
+> **Language (read FIRST)**: Before any status output, read `prefs.global.outputLanguage` from `$HOME/.claude/multi-agent-preferences.json` and render every conversational line in it. `AskUserQuestion` and external payloads stay English. Full contract: `$HOME/.claude/multi-agent-refs/rules.md` "Language Application".
 
 ## Canonical Invocation (per CLI)
 
@@ -34,7 +34,7 @@ the canonical invocations and the only forms surfaced in autocomplete.
 
 ## Input Parsing
 
-Classification schema lives in `commands/multi-agent/_input-parser.md`  -  canonical source. Summary:
+Classification schema lives in `$HOME/.claude/multi-agent-refs/_input-parser.md`  -  canonical source. Summary:
 
 | # | Pattern | Type | Picker flow |
 |---|---------|------|--------|
@@ -65,8 +65,8 @@ Lib scripts (`~/.claude/lib/`):
 | `https://{JIRA_HOST}/browse/{JIRA_KEY}-XXXXX` | Jira URL  -  fetch, select project, start pipeline |
 | `{JIRA_KEY}-XXXXX` | Jira ID  -  fetch, select project, start pipeline |
 | Free-text | Bug/feature description  -  select project, start pipeline |
-| `jira` | 4-step picker (hesap → proje → issue → dev-context)  -  `commands/multi-agent/jira.md` delegate |
-| `issue` | 4-step picker (hesap → repo → issue → dev-context)  -  `commands/multi-agent/issue.md` delegate |
+| `jira` | 4-step picker (hesap → proje → issue → dev-context)  -  `commands/multi-agent/jira/SKILL.md` delegate |
+| `issue` | 4-step picker (hesap → repo → issue → dev-context)  -  `commands/multi-agent/issue/SKILL.md` delegate |
 | `status` | List all tasks (scan `$HOME/.claude/logs/multi-agent/`) |
 | `log [id]` | Show task log |
 | `resume [id]` | Resume paused task |
@@ -78,7 +78,7 @@ Lib scripts (`~/.claude/lib/`):
 | `build-optimize` | iOS-only Xcode build performance wrapper. Vendored `xcode-build-orchestrator`'a dispatch eder; benchmark + compilation / project / SPM analyzer'lar + recommend-first plan `.build-benchmark/optimization-plan.md`. Non-iOS stack'lerde fail-fast. |
 | `channels [PR-url\|#N\|Jira-url\|Jira-id] [--channels pr,jira,confluence,wiki] [--content normal,test,auto-diff,note] [--message "..."]` | Post task report to multi-select channels (PR description, Jira comment, Confluence page, Wiki pages) with multi-select content sources. Humanizer pass per-channel. Bitbucket PR updates use reviewer-preserving PUT. Phase 7 delegates to this command; also invocable post-hoc for fixes made outside the pipeline. No worktree. |
 | `generate ["desc"] [figma-url] [swagger-url]` | Create a standards-compliant Jira issue: asks the type (**Task** / **Bug** / **Story**), mines the project's recent same-type issues for conventions (summary format, labels, priority, test-scenario style), detects the active sprint, drafts from a standard template with auto-sizing sections (Design Reference / API Contract / Screenshots appear only when their source is given), asks about unknown fields, then full draft preview + explicit approval before create. No worktree, no commits. |
-| `test` or `test [args]` | UI Bug Hunter  -  screenshot + tap + analyze on booted simulator via MCP (read `$HOME/.claude/commands/sim-test.md`). `/multi-agent:test` also resolves here via the `commands/multi-agent/test.md` delegate. |
+| `test` or `test [args]` | UI Bug Hunter  -  screenshot + tap + analyze on booted simulator via MCP (read `$HOME/.claude/commands/sim-test.md`). `/multi-agent:test` also resolves here via the `commands/multi-agent/test/SKILL.md` delegate. |
 | `manual-test [#id]` | Phase 5 standalone Manual Test  -  checks out the task branch, prints Xcode / SourceTree hints, waits for user verdict (`ok` / `fix: ...`). |
 | `stack [ios\|android\|backend\|mobile\|all]` | Swap skills for next conversation. No arg = show current stack. |
 | `language [en\|tr]` | Show or set the assistant `outputLanguage` (explanations and chat replies). `promptLanguage` is locked to `en` and is not toggleable. No arg = show current `outputLanguage`. With `en` or `tr` = set and persist `outputLanguage`. External payloads (commits, PR bodies, Jira) stay English. |
@@ -97,38 +97,38 @@ This command uses lazy loading for token efficiency. Read the relevant sub-file
 
 **File layout:**
 - `commands/multi-agent/*.md` → **invocable actions** (each gets its own `/multi-agent:<name>` slash command)
-- `commands/multi-agent/refs/**` → **internal references**, read by the main command, never invoked directly (surface as `/multi-agent:refs:...` in autocomplete  -  the prefix signals "not an action")
+- `$HOME/.claude/multi-agent-refs/**` → **internal references**, read by the main command, never invoked directly (surface as `/multi-agent:refs:...` in autocomplete  -  the prefix signals "not an action")
 
 | Route | File to Read |
 |-------|-------------|
-| Pipeline task (any input type) | `$HOME/.claude/commands/multi-agent/refs/phases.md` + `commands/multi-agent/_account-picker.md` + `_repo-picker.md` + `_dev-context.md` |
-| `jira` | `$HOME/.claude/commands/multi-agent/jira.md` (4-step picker) |
-| `issue` | `$HOME/.claude/commands/multi-agent/issue.md` (4-step picker) |
-| Direct ID (`{JIRA_KEY}-N`, `#N`, gh URL, repo#N, free-text) | `commands/multi-agent/_input-parser.md` → `_account-picker.md` → `lib/issue-fetcher.sh` (maturity check) → `_dev-context.md` |
-| `setup` | `$HOME/.claude/commands/multi-agent/setup.md` |
-| `help` | `$HOME/.claude/commands/multi-agent/help.md` |
-| `channels` | `$HOME/.claude/commands/multi-agent/channels.md` |
-| `sync` | `$HOME/.claude/commands/multi-agent/sync.md` |
+| Pipeline task (any input type) | `$HOME/.claude/multi-agent-refs/phases.md` + `$HOME/.claude/multi-agent-refs/_account-picker.md` + `_repo-picker.md` + `_dev-context.md` |
+| `jira` | `$HOME/.claude/commands/multi-agent/jira/SKILL.md` (4-step picker) |
+| `issue` | `$HOME/.claude/commands/multi-agent/issue/SKILL.md` (4-step picker) |
+| Direct ID (`{JIRA_KEY}-N`, `#N`, gh URL, repo#N, free-text) | `$HOME/.claude/multi-agent-refs/_input-parser.md` → `_account-picker.md` → `lib/issue-fetcher.sh` (maturity check) → `_dev-context.md` |
+| `setup` | `$HOME/.claude/commands/multi-agent/setup/SKILL.md` |
+| `help` | `$HOME/.claude/commands/multi-agent/help/SKILL.md` |
+| `channels` | `$HOME/.claude/commands/multi-agent/channels/SKILL.md` |
+| `sync` | `$HOME/.claude/commands/multi-agent/sync/SKILL.md` |
 | `clear-logs` | Handled inline  -  scan + delete agent-log.md/agent-state.json files |
-| `review` | `$HOME/.claude/commands/multi-agent/review.md` |
-| `analysis` | `$HOME/.claude/commands/multi-agent/analysis.md` |
-| `build-optimize` | `$HOME/.claude/commands/multi-agent/build-optimize.md` |
-| `local` | `$HOME/.claude/commands/multi-agent/local.md` |
-| `local-autopilot` | `$HOME/.claude/commands/multi-agent/local-autopilot.md` |
-| `dev-local` | `$HOME/.claude/commands/multi-agent/dev-local.md` |
-| `dev-local-autopilot` | `$HOME/.claude/commands/multi-agent/dev-local-autopilot.md` |
-| `generate` | `$HOME/.claude/commands/multi-agent/generate.md` (loads `refs/generate-issue.md`) |
-| `stack` | `$HOME/.claude/commands/multi-agent/stack.md` |
+| `review` | `$HOME/.claude/commands/multi-agent/review/SKILL.md` |
+| `analysis` | `$HOME/.claude/commands/multi-agent/analysis/SKILL.md` |
+| `build-optimize` | `$HOME/.claude/commands/multi-agent/build-optimize/SKILL.md` |
+| `local` | `$HOME/.claude/commands/multi-agent/local/SKILL.md` |
+| `local-autopilot` | `$HOME/.claude/commands/multi-agent/local-autopilot/SKILL.md` |
+| `dev-local` | `$HOME/.claude/commands/multi-agent/dev-local/SKILL.md` |
+| `dev-local-autopilot` | `$HOME/.claude/commands/multi-agent/dev-local-autopilot/SKILL.md` |
+| `generate` | `$HOME/.claude/commands/multi-agent/generate/SKILL.md` (loads `$HOME/.claude/multi-agent-refs/generate-issue.md`) |
+| `stack` | `$HOME/.claude/commands/multi-agent/stack/SKILL.md` |
 | `language` | Handled inline  -  set/show prompt language in preferences |
-| SwiftUI component task (iOS) | `$HOME/.claude/commands/multi-agent/refs/swiftui-guide.md` |
-| Jetpack Compose task (Android) | `$HOME/.claude/commands/multi-agent/refs/android-guide.md` |
-| Backend API task | `$HOME/.claude/commands/multi-agent/refs/backend-guide.md` |
-| Frontend component task | `$HOME/.claude/commands/multi-agent/refs/frontend-guide.md` |
-| Phase 1 or Phase 7 (knowledge) | `$HOME/.claude/commands/multi-agent/refs/knowledge.md` |
-| Token lookup needed | `$HOME/.claude/commands/multi-agent/refs/keychain.md` |
-| Audit tools (Phase 5/6) | `$HOME/.claude/commands/multi-agent/refs/audit-guide.md` |
-| `test` | `$HOME/.claude/commands/sim-test.md` (colon-form `/multi-agent:test` uses the delegate at `commands/multi-agent/test.md`) |
-| `manual-test` | `$HOME/.claude/commands/multi-agent/manual-test.md` |
+| SwiftUI component task (iOS) | `$HOME/.claude/multi-agent-refs/swiftui-guide.md` |
+| Jetpack Compose task (Android) | `$HOME/.claude/multi-agent-refs/android-guide.md` |
+| Backend API task | `$HOME/.claude/multi-agent-refs/backend-guide.md` |
+| Frontend component task | `$HOME/.claude/multi-agent-refs/frontend-guide.md` |
+| Phase 1 or Phase 7 (knowledge) | `$HOME/.claude/multi-agent-refs/knowledge.md` |
+| Token lookup needed | `$HOME/.claude/multi-agent-refs/keychain.md` |
+| Audit tools (Phase 5/6) | `$HOME/.claude/multi-agent-refs/audit-guide.md` |
+| `test` | `$HOME/.claude/commands/sim-test.md` (colon-form `/multi-agent:test` uses the delegate at `commands/multi-agent/test/SKILL.md`) |
+| `manual-test` | `$HOME/.claude/commands/multi-agent/manual-test/SKILL.md` |
 
 **Modifier flags** (`--dev`, `--local`, `autopilot`) and **ops** (`status`, `log`, `resume`, `kill`, `clear-logs`, `purge`, `review`) are parsed inline by this file  -  no separate spec files, they compose with the pipeline or do one-shot work.
 
@@ -149,19 +149,19 @@ This command uses lazy loading for token efficiency. Read the relevant sub-file
 - `$HOME/.claude/agents/backend-architect.md`  -  Backend/API architecture review (`preferredModel: fable`)
 - `$HOME/.claude/agents/security-auditor.md`  -  Security audit (`preferredModel: opus`)
 
-**Per-persona model routing:** Before each Agent dispatch, the orchestrator reads `preferredModel` from the persona file and exports `CLAUDE_CODE_SUBAGENT_MODEL` (Claude Code) / passes `--model` (Copilot CLI). Precedence: per-dispatch `PHASE_MODEL_OVERRIDE` > persona `preferredModel` > `fable` (falls back per `refs/features/model-fallback.md`). Full contract: `skills/shared/core/multi-agent/SKILL.md#agent-dispatch--per-persona-model-routing-v610`.
+**Per-persona model routing:** Before each Agent dispatch, the orchestrator reads `preferredModel` from the persona file and exports `CLAUDE_CODE_SUBAGENT_MODEL` (Claude Code) / passes `--model` (Copilot CLI). Precedence: per-dispatch `PHASE_MODEL_OVERRIDE` > persona `preferredModel` > `fable` (falls back per `$HOME/.claude/multi-agent-refs/features/model-fallback.md`). Full contract: `skills/shared/core/multi-agent/SKILL.md#agent-dispatch--per-persona-model-routing-v610`.
 
 ---
 
 ## language Command
 
-Toggles only `outputLanguage` (`promptLanguage` is locked to `"en"`  -  see `refs/rules.md` section "Language Application"). Full spec: `pipeline/commands/multi-agent/language.md`.
+Toggles only `outputLanguage` (`promptLanguage` is locked to `"en"`  -  see `$HOME/.claude/multi-agent-refs/rules.md` section "Language Application"). Full spec: `pipeline/commands/multi-agent/language/SKILL.md`.
 
 ---
 
 ## jira Command
 
-Delegates to `commands/multi-agent/jira.md` (4-step picker). Read that file, then apply:
+Delegates to `commands/multi-agent/jira/SKILL.md` (4-step picker). Read that file, then apply:
 
 1. **Pick account**  -  `_account-picker.md` (`lib/account-resolver.sh --providers jira`)
 2. **Pick projects (multi)**  -  Jira project list, multi-select
@@ -176,7 +176,7 @@ Steps 1-5 are detailed in `jira.md`; the Interactive Launcher (below) consumes t
 
 ## issue Command
 
-Delegates to `commands/multi-agent/issue.md` (4-step picker). Read that file, then apply:
+Delegates to `commands/multi-agent/issue/SKILL.md` (4-step picker). Read that file, then apply:
 
 1. **Pick account**  -  `_account-picker.md` (`--providers github`)
 2. **Pick repos (multi)**  -  `lib/repo-cache.sh github <owner>`, multi-select
@@ -257,7 +257,7 @@ When called with `review`:
 
 ## Rules
 
-**Non-negotiable rules that apply to every phase, every mode.** Full reference at `$HOME/.claude/commands/multi-agent/refs/rules.md`  -  read before any invocation that writes code, commits, or posts to external systems.
+**Non-negotiable rules that apply to every phase, every mode.** Full reference at `$HOME/.claude/multi-agent-refs/rules.md`  -  read before any invocation that writes code, commits, or posts to external systems.
 
 Quick pointers:
 - No "Copilot/AI/generated by" attribution. Ever.
@@ -269,7 +269,7 @@ Quick pointers:
 - Branch: `feature/` or `bugfix/` prefix.
 - 3-iteration hard kill on any retry loop.
 
-Load `refs/rules.md` for the full contract (Swift-specific rules, reviewer policy, secret scan, subagent contract).
+Load `$HOME/.claude/multi-agent-refs/rules.md` for the full contract (Swift-specific rules, reviewer policy, secret scan, subagent contract).
 
 ---
 

package/pipeline/commands/multi-agent/{analysis.md → analysis/SKILL.md}

@@ -11,7 +11,7 @@ This command is **independent** from the orchestrator's Phase 1 analysis (which
 
 **Scope (development analysis, not a screen spec)**: this command produces a *development* analysis  -  what to build, the architecture plan, files to add, the test plan, and which existing components to reuse. It is NOT a design / screen-anatomy specification. The design itself is already produced elsewhere (figma-to-swiftui builds the components and binds them via Code Connect), so the analysis consumes those existing bindings as the source of truth for "what already exists" (see Phase 1b.1) instead of re-deriving the design. Do not grow this command toward exhaustive per-screen layout / gesture / state documentation; keep it development-focused.
 
-> **Language**: Per `refs/rules.md` Language Application matrix - instruction prose stays English (this file is read as a system prompt). `AskUserQuestion.label` and `header` stay English. `question` and `description` follow `outputLanguage`. The emitted analysis document body follows `outputLanguage` (`tr` or `en`).
+> **Language**: Per `$HOME/.claude/multi-agent-refs/rules.md` Language Application matrix - instruction prose stays English (this file is read as a system prompt). `AskUserQuestion.label` and `header` stay English. `question` and `description` follow `outputLanguage`. The emitted analysis document body follows `outputLanguage` (`tr` or `en`).
 
 ## Locked decisions (do not re-ask)
 
@@ -41,7 +41,7 @@ When citing a Locked decision in code or docs, prefer `Locked <n> (<short label>
 6. **No auto-commit.** Even after a successful local write or Confluence post, the command never runs `git add` or `git commit`. The user does that themselves if they want.
 7. **Humanizer punctuation policy is non-negotiable.** No em-dash, en-dash, ellipsis, curly quotes, or section sign in any emitted text. Verification grep is run before the file lands.
 8. **Standards binding.** When `evidence.standards[]` is non-empty, Section 7 architectural decisions must cite their binding source by file plus section. Decisions that contradict a binding source go to Section 7 Risks rather than silently overriding.
-9. **Per-platform output split.** One markdown file is produced per selected platform under `analysis/<feature>-<platform>.md`. There is no merged single file. Section duplication follows the A3 hybrid table in `refs/analysis-template.md` (Sections 1 + 4 duplicated verbatim; Sections 2, 3, 5, 6, 7 projected per platform). Each file carries a YAML front-matter header naming the feature, platform, generated-at timestamp, sibling file list, and binding standards source.
+9. **Per-platform output split.** One markdown file is produced per selected platform under `analysis/<feature>-<platform>.md`. There is no merged single file. Section duplication follows the A3 hybrid table in `$HOME/.claude/multi-agent-refs/analysis-template.md` (Sections 1 + 4 duplicated verbatim; Sections 2, 3, 5, 6, 7 projected per platform). Each file carries a YAML front-matter header naming the feature, platform, generated-at timestamp, sibling file list, and binding standards source.
 10. **Output destination is asked after drafts exist.** Phase 3 renders each per-platform markdown to `/tmp/analysis-<feature-slug>-<timestamp>/` first. Only then does Phase 3.5 surface the output-destination picker (Local / Confluence / Jira). Drafts in `/tmp/` are the source of truth on resume; `/multi-agent:resume` re-uses them when `phase == awaiting_output_decision`.
 11. **Repo-evidence reuse-first.** Phase 1b runs a repo-evidence collector against each selected repo, producing 13 buckets (services, dtos, useCases, validationRules, domainEntities, routes, coordinators, diConfigurators, uiComponents, tokens, localizationKeys, testingIdentifiers, analyticsEvents) with each row tagged `direct-match | same-domain | cross-cutting`. Section 7 emits `Reuse existing X (file:line)` rows for `direct-match` items and advisory rows for `cross-cutting` items. New-write tasks for items that have a `direct-match` are downgraded to Risks ("existing X candidate found; reuse or document why a new one is needed").
 12. **MUST: Figma access - 3-tier fallback chain, pipeline-wide.** When any Figma URL or node ID is supplied (Phase 0 Step 5), Phase 1 establishes a Figma ground-truth artefact via the 3-tier chain before any UI synthesis runs: (1) Figma MCP `get_design_context` / `get_screenshot` / `get_metadata` for every referenced frame, with one re-auth retry on auth failure; (2) Figma REST API (`GET /v1/files/{fileKey}/nodes`, `GET /v1/images/{fileKey}`) using the PAT resolved through `~/.claude/lib/credential-store.sh get <logical-key>` where `<logical-key>` = `prefs.global.keychainMapping.figma_pat`; (3) a user-attached screenshot as last resort. The tier in use is persisted as `state.figmaAccess.tier`. On Tier 1 the `CodeConnectSnippet` blocks name the exact target component consumed verbatim in Section 2 and Section 7. On Tier 2 the canonical-component decision falls back to the repo's `*.figma.swift` / `*.figma.kt` mappings keyed by `fileKey` + `nodeId`. On Tier 3 the canonical-component decision becomes "tier-3 best-fit pending design review" and produces a forced Open Question in Section 7 plus a Phase 4 `review_blocking` flag. Sound-alike alternatives, "more flexible" wrappers, and extrapolating from Confluence text-only "Component Kompozisyonu" / "Component Inventory" tables are forbidden in every tier. If all three tiers fail, halt the run and ask the user for access; never proceed with text-derived guesses. Section 7 architecture decisions must cite the node ID (Tier 1 or Tier 2) or the user-screenshot reference (Tier 3) for every UI atom row. Violations cost rebuild rounds (raw-primitive substitution; sound-alike-component swap). Generic rule rationale and checklist: see `pipeline/rules/figma-pipeline.md` "MUST: Figma access - 3-tier fallback chain (BLOCKING, pipeline-wide)". Memory: `[[figma-no-guesswork]]`.
@@ -54,7 +54,7 @@ When citing a Locked decision in code or docs, prefer `Locked <n> (<short label>
 19. **All Figma variants drilled.** When a Figma section URL is supplied, all child frames are drilled, not just the canonical default. The renderer enumerates child frames via `mcp__claude_ai_Figma__get_metadata` (Tier 1) or `figma-screenshot.sh --section` (Tier 2) and produces one Section 5.1 row per child frame.
 20. **Localization 8-locale full coverage.** Section 10 new keys carry filled cells for all eight supported locales: TR, EN, AR, DE, ES, FR, RU, ZH. Placeholder values `[bekleniyor: çeviri ekibi]` / `[pending: translation team]` are accepted as a soft state and the dispatch report flags `i18n_pending: <count>` as a blocker. Empty cells are rejected outright.
 21. **References at the bottom.** The user-supplied input table is rendered as Section 21 References, not as Section 1. Readers see goals, flow, and design before sources.
-22. **Platform-agnostic template + Pass B render.** The template (`refs/analysis-template.md`) describes concepts (state holder, view, navigator, use case, repository, DTO, state model, DI register, localization key, accessibility identifier, test method) without platform-specific class names. Pass B (Phase 2b) projects each concept onto the selected platform using conventions extracted at Phase 1c.
+22. **Platform-agnostic template + Pass B render.** The template (`$HOME/.claude/multi-agent-refs/analysis-template.md`) describes concepts (state holder, view, navigator, use case, repository, DTO, state model, DI register, localization key, accessibility identifier, test method) without platform-specific class names. Pass B (Phase 2b) projects each concept onto the selected platform using conventions extracted at Phase 1c.
 23. **Convention extraction mandatory (Phase 1c).** After Phase 1b repo evidence, Phase 1c extracts seven pattern groups (folder structure, class naming, UI state model, test method naming, accessibility identifier, localization key, DI registration) per selected repo via `~/.claude/lib/extract-conventions.sh`. Output lands in `state.analysisSpec.evidence.conventions[<repo>]` with confidence levels (high / medium / low / none) and evidence file citations.
 24. **Pass B cell footnote mandatory.** Every cell Pass B fills in Section 13 (Architecture Plan concept table) and any other per-platform projection carries a footnote of the form `^[<convention-key> <confidence>: <evidence-source>]` pointing back to Phase 1c output. Cells with `confidence: low | none` cite `conventions-defaults.md:C<n>-<platform>` and emit a row in Section 20 Risks ("convention fallback applied"). Footnote-less cells fail the dispatch gate.
 25. **Lite mode for small features.** When Phase 1 signals indicate a small feature, Lite mode auto-activates and renders only Sections 1, 2, 4, 9, 13, 14, 21, plus optional 23 Changelog. The user can force Lite with `--lite` or force Full with `--full`. Full mode is the default for new feature analyses. **Scoring (v9.1.0+):** three independent signals - `confluenceSpecLines < 100`, `figmaFramesCount <= 1`, `repoDirectMatchCount >= 8`. Each true signal scores 1 point; Lite auto-activates at `score >= 2`. Previous AND-threshold (`v8.12.0..v9.0.x`) was too strict and forced small features into Full mode when one signal was just over the line. Explicit user flags (`--lite` / `--full`) always win over auto-scoring.
@@ -70,7 +70,7 @@ When citing a Locked decision in code or docs, prefer `Locked <n> (<short label>
 
 ## Template
 
-v3 template at `refs/analysis-template.md`. Full mode renders 23 main sections + 3 footer sections (Glossary, Changelog, References-as-footer per Locked 21). Lite mode renders 7 main sections (1 Summary, 2 Goals + Non-Goals, 4 User Stories, 9 API Contracts, 13 Architecture, 14 Files to Add, 21 References) plus optional 23 Changelog. Section omission rule (Locked 2): zero-evidence sections drop entirely; numbering re-flows to stay sequential `1..N` over the rendered set.
+v3 template at `$HOME/.claude/multi-agent-refs/analysis-template.md`. Full mode renders 23 main sections + 3 footer sections (Glossary, Changelog, References-as-footer per Locked 21). Lite mode renders 7 main sections (1 Summary, 2 Goals + Non-Goals, 4 User Stories, 9 API Contracts, 13 Architecture, 14 Files to Add, 21 References) plus optional 23 Changelog. Section omission rule (Locked 2): zero-evidence sections drop entirely; numbering re-flows to stay sequential `1..N` over the rendered set.
 
 Numbering is plain Arabic numerals: `## 1. Geliştirme Özeti` / `## 1. Summary`, `## 2. Hedefler ve Karşı Hedefler` / `## 2. Goals and Non-Goals`. No special section characters; the section sign codepoint (U+00A7) is banned by `feedback_no-section-sign`.
 
@@ -82,7 +82,7 @@ The template is platform-agnostic (Locked 22). It speaks in concepts (state hold
 
 Sequential `AskUserQuestion` chain. Each answer is written to state under `state.analysisSpec.*`.
 
-**Step narration (required, per `refs/picker-contract.md`)**: the chain length is known up front  -  1 analysis-name + 1 account + 1 platform + 1 repo-round per selected platform + 1 input-URL batch (so a single-platform run is 5 steps; account is skipped for local-only flows, which lowers the total). Before each step's `AskUserQuestion`, print the narrator line `<localized: "Step <i>/<n>: <what this step decides>">` in `outputLanguage`. Auto-resolved steps (single account, local-only) still print their breadcrumb with the resolution noted. This is what makes the picker show, step by step, what it is doing.
+**Step narration (required, per `$HOME/.claude/multi-agent-refs/picker-contract.md`)**: the chain length is known up front  -  1 analysis-name + 1 account + 1 platform + 1 repo-round per selected platform + 1 input-URL batch (so a single-platform run is 5 steps; account is skipped for local-only flows, which lowers the total). Before each step's `AskUserQuestion`, print the narrator line `<localized: "Step <i>/<n>: <what this step decides>">` in `outputLanguage`. Auto-resolved steps (single account, local-only) still print their breadcrumb with the resolution noted. This is what makes the picker show, step by step, what it is doing.
 
 #### Step 0 - Language resolution (BLOCKING, runs before any picker)
 
@@ -364,7 +364,7 @@ Each field has shape `{ pattern, example, confidence, evidenceFiles, alternative
 
 Phase 2 Section 20 Risks reads this list and emits one open question per entry.
 
-**Fallback source**: when `confidence == "none"` AND `evidence.standards[]` does not contain an explicit rule for that field, the renderer reads `~/.claude/commands/multi-agent/refs/conventions-defaults.md` and applies the platform default.
+**Fallback source**: when `confidence == "none"` AND `evidence.standards[]` does not contain an explicit rule for that field, the renderer reads `$HOME/.claude/multi-agent-refs/conventions-defaults.md` and applies the platform default.
 
 **Caching (Locked 27)**: compute `evidence_digest = sha256(featureName || sorted(platforms) || hash(evidence.repoEvidence) || hash(evidence.conventions))`. Cache key on disk: `/tmp/multi-agent-analysis-cache/<digest>.json` with mtime <= 24h. Cache hit skips Phase 1b and Phase 1c. `--no-cache` flag forces re-run.
 
@@ -402,7 +402,7 @@ synthesizedSections = {
 |---------|------------------|
 | 1. Scope | Always present. Derive in-scope / out-of-scope, audience from Jira summary + Confluence title + feature name. LLM inference allowed. Shared across all per-platform files verbatim. |
 | 2. Design | If `evidence.figma[]` is empty AND no `evidence.repoEvidence[*].buckets.uiComponents` direct-match → `byPlatform[*]` = null. Otherwise produce frame gallery (shared) + per-platform component inventory (rows from `evidence.repoEvidence[<repo>].buckets.uiComponents`). |
-| 3. Localization | If both Figma text layers and `evidence.repoEvidence[*].buckets.localizationKeys` are empty → `byPlatform[*]` = null. Otherwise produce the key table; per-platform `Source` column derives from the platform's localization file convention. Reuse 8-locale logic from `pipeline/skills/figma-ios/figma-to-component/phases/phase-2b-localization.md`. |
+| 3. Localization | If both Figma text layers and `evidence.repoEvidence[*].buckets.localizationKeys` are empty → `byPlatform[*]` = null. Otherwise produce the key table; per-platform `Source` column derives from the platform's localization file convention. Reuse the 8-locale set (ar, de, en, es, fr, it, ru, tr) with standard localization-key naming (RTL for ar). |
 | 4. API Contracts | If `evidence.swagger[]` is empty AND no Confluence embedded API table AND no `evidence.repoEvidence[*].buckets.services` direct-match → null. Otherwise produce endpoint summary + per-endpoint tables (shared verbatim across files). |
 | 5. Deeplink / Push | Per-platform: iOS uses Universal Links / `UNUserNotificationCenter`; Android uses `intent-filter` / FCM; Frontend uses web URL routing; Backend file omits this section entirely. |
 | 6. Business + Tests | If sections 2 and 4 are both `null` AND `evidence.firebase[]` is empty → null. Otherwise produce use-case + mock stubs + per-platform test skeletons + shared Firebase events table. Reuse Red-Green-Refactor naming from `pipeline/rules/tdd.md`. |
@@ -464,7 +464,7 @@ For each `platform` in `state.analysisSpec.platforms[]`:
    - `frontend` → `evidence.standards[]` entries matching `react`, `vue`, `next`, `sveltekit` → `~/.claude/rules/code-style.md`
 2. **Apply per-platform omission rules.** Backend-only file drops Sections 5, 6, 7, 8, 16. Frontend with no UI inventory still keeps 5 (UI exists in code). Sections 1, 2, 4, 9, 13, 14, 20, 21 always present per Locked decision 2 + 13.
 3. **Resolve mode.** If user passed `--lite` → Lite. If user passed `--full` → Full. Otherwise use `state.analysisSpec.liteModeAuto`. Lite mode renders only Sections 1, 2, 4, 9, 13, 14, 21 plus optional 23.
-4. **Produce YAML front-matter header** (see `refs/analysis-template.md`). Include `mode: full | lite`.
+4. **Produce YAML front-matter header** (see `$HOME/.claude/multi-agent-refs/analysis-template.md`). Include `mode: full | lite`.
 5. **Read conventions for this platform's repo.** For each cell Pass B fills in Section 13 and in any per-platform projection (Sections 5, 6, 7, 8, 10, 11, 13, 14, 15, 16, 17), read `state.analysisSpec.evidence.conventions[<repo>].<field>` and emit the value with a footnote (Locked 24). If `conventionOverrides` has an entry for that field, use the override and footnote with `^[user-override: <reason>]` instead of evidence path.
 6. **Concatenate non-null sections in canonical order.** Numbering stays sequential `1..N` over the rendered set (omitted sections do not create gaps).
 7. **Schema validation** on the per-platform spec object:
@@ -475,7 +475,7 @@ For each `platform` in `state.analysisSpec.platforms[]`:
 
 ### Phase 3 - Humanize & buffer drafts (no side effects)
 
-1. **Language resolution**: read `prefs.global.outputLanguage` (`tr` or `en`, default `tr`). Use the `Output language matrix` table in `refs/analysis-template.md` to swap headings and system strings.
+1. **Language resolution**: read `prefs.global.outputLanguage` (`tr` or `en`, default `tr`). Use the `Output language matrix` table in `$HOME/.claude/multi-agent-refs/analysis-template.md` to swap headings and system strings.
 
 2. **Per-platform markdown render**: for each platform in `state.analysisSpec.platforms[]`, concatenate the per-platform spec into one markdown file. Tables in pipe-syntax. Numbering uses plain `## 1.`, `## 2.`, ... - omitted sections do **not** create gaps. Visible numbering is sequential 1..N over the rendered set.
 
@@ -528,8 +528,8 @@ Iterate `state.analysisSpec.outputs.requested`. For each target:
 | Target | Action |
 |--------|--------|
 | Local | For each per-platform draft, `cp /tmp/analysis-<feature-slug>-<ts>/<feature>-<platform>.md` into `analysis/<feature>-<platform>.md` in the matching repo's working tree. When multiple repos exist for the same platform, the file is duplicated into each and the dispatch report lists every destination. **No commit.** |
-| Confluence | Re-humanize each per-platform draft with `formal-stakeholder` tone. One Confluence page per platform under the chosen parent, titled `<Feature> - <Platform>`. Cross-link siblings inside each page via `<ac:link><ri:page ri:content-title="<Feature> - <OtherPlatform>"/></ac:link>`. Markdown -> storage XML via `pipeline/commands/multi-agent/refs/channels/confluence.md`. Re-emit on existing pages uses PUT with version bump. |
-| Jira | Re-humanize the combined body with `informal-technical` tone. Concatenate per-platform drafts under `## Platform: iOS`, `## Platform: Android`, `## Platform: Backend`, `## Platform: Frontend` separators (in the order platforms were selected). PUT the chosen issue's description via `pipeline/commands/multi-agent/refs/channels/jira.md`. |
+| Confluence | Re-humanize each per-platform draft with `formal-stakeholder` tone. One Confluence page per platform under the chosen parent, titled `<Feature> - <Platform>`. Cross-link siblings inside each page via `<ac:link><ri:page ri:content-title="<Feature> - <OtherPlatform>"/></ac:link>`. Markdown -> storage XML via `$HOME/.claude/multi-agent-refs/channels/confluence.md`. Re-emit on existing pages uses PUT with version bump. |
+| Jira | Re-humanize the combined body with `informal-technical` tone. Concatenate per-platform drafts under `## Platform: iOS`, `## Platform: Android`, `## Platform: Backend`, `## Platform: Frontend` separators (in the order platforms were selected). PUT the chosen issue's description via `$HOME/.claude/multi-agent-refs/channels/jira.md`. |
 
 **Output capture**: fill `state.analysisSpec.outputs.localPaths[]` (one entry per per-platform-per-repo write), `outputs.confluencePageUrls[]` (one entry per platform), `outputs.jiraIssueKey` (single string).
 
@@ -627,12 +627,12 @@ When `phase == "cancelled_at_pass_b_preview"`:
 | `~/.claude/lib/figma-screenshot.sh` | Phase 2b Tier 2 Figma image downloader (REST API, section drill, 2x scale PNG, manifest.json) |
 | `~/.claude/lib/md2confluence-v3.py` | Phase 4 Confluence dispatch (multipart attachments, `<ac:image>` injection, mermaid macro + fallback, tooltip footnote macro, punctuation gate) |
 | `pipeline/skills/shared/external/humanizer/SKILL.md` | Phase 3 tone pass |
-| `pipeline/skills/figma-ios/figma-to-component/phases/phase-2b-localization.md` | Section 10 localization generation |
-| `pipeline/commands/multi-agent/refs/channels/confluence.md` | Phase 4 Confluence dispatch |
-| `pipeline/commands/multi-agent/refs/channels/jira.md` | Phase 4 Jira dispatch |
+| 8-locale set (ar, de, en, es, fr, it, ru, tr) + localization-key naming (inline) | Section 10 localization generation |
+| `$HOME/.claude/multi-agent-refs/channels/confluence.md` | Phase 4 Confluence dispatch |
+| `$HOME/.claude/multi-agent-refs/channels/jira.md` | Phase 4 Jira dispatch |
 | `pipeline/rules/tdd.md` | Section 15 test naming |
-| `pipeline/commands/multi-agent/refs/analysis-template.md` | Template master copy + language matrix (v3 - 23 main sections + 3 footer) |
-| `pipeline/commands/multi-agent/refs/conventions-defaults.md` | Pass B fallback defaults (4 platforms x 7 pattern groups) - applied when convention confidence is none AND standards binding is silent |
+| `$HOME/.claude/multi-agent-refs/analysis-template.md` | Template master copy + language matrix (v3 - 23 main sections + 3 footer) |
+| `$HOME/.claude/multi-agent-refs/conventions-defaults.md` | Pass B fallback defaults (4 platforms x 7 pattern groups) - applied when convention confidence is none AND standards binding is silent |
 | `~/.claude/skills/checkin-confluence-mirror/sync.py` | Confluence-embedded API table parser pattern |
 | `~/<project>-Standards.md` | Canonical home-dir standards reference (auto-detected at Q5 option 2; exact filename from `prefs.projects[<project>].standardsFile`) |
 | `~/.claude/rules/*.md` | Fallback rules when `evidence.standards[]` is empty |
@@ -641,9 +641,9 @@ When `phase == "cancelled_at_pass_b_preview"`:
 
 Per the `analysis-output-confluence-on-request` memory, Confluence post is NEVER default-selected at Phase 3.5 (the output destination picker). Only when the user explicitly asks ("post this to Confluence", "create a page under the Coding Documentation parent", etc.) does the command:
 
-1. Resolve the token via `~/.claude/lib/credential-store.sh get <key>`, where `<key>` is read from `prefs.global.keychainMapping.confluence` (per-user mapping; never hardcode the service name in this doc - see channel adapter doc `pipeline/commands/multi-agent/refs/channels/confluence.md` for the lookup contract).
+1. Resolve the token via `~/.claude/lib/credential-store.sh get <key>`, where `<key>` is read from `prefs.global.keychainMapping.confluence` (per-user mapping; never hardcode the service name in this doc - see channel adapter doc `$HOME/.claude/multi-agent-refs/channels/confluence.md` for the lookup contract).
 2. Use parent page URL from user input. No default parent is hardcoded here; the user picks one at the prompt (LRU recents come from `prefs.projects[<project>].confluenceUrls`).
-3. Convert markdown to storage XML using the table in `pipeline/commands/multi-agent/refs/channels/confluence.md`.
+3. Convert markdown to storage XML using the table in `$HOME/.claude/multi-agent-refs/channels/confluence.md`.
 4. Upload Figma frame screenshots as page attachments via `POST /rest/api/content/{pageId}/child/attachment` (cache the MCP asset locally first because the upstream URLs expire after 7 days).
 5. Reference attachments inside the page body via `<ac:image><ri:attachment ri:filename="frame-<nodeId>.png"/></ac:image>`.
 6. POST `/rest/api/content` to create the page (or PUT with version bump when updating). Surface the resulting page URL in the Phase 4 report.

package/pipeline/commands/multi-agent/{analysis-resolve.md → analysis-resolve/SKILL.md}

@@ -5,11 +5,11 @@ argument-hint: "[path/to/analysis/<feature>-<platform>.md] [--autonomous]"
 
 # multi-agent analysis-resolve - Open Question Resolver for Analysis Docs
 
-Companion command to `/multi-agent:analysis`. Takes an `analysis/<feature>-<platform>.md` (template v3) and walks the user through the open rows of Section 20 (Riskler ve Acik Sorular / Risks and Open Questions) one row at a time, proposing source-labeled answer candidates and merging the chosen answer into the proper body section. The pattern is ported from the ai-mobile-toolkit resolver skills (`mobile-screen-spec-resolver`, `mobile-flow-spec-resolver`) and adapted to the analysis doc contract.
+Companion command to `/multi-agent:analysis`. Takes an `analysis/<feature>-<platform>.md` (template v3) and walks the user through the open rows of Section 20 (Riskler ve Acik Sorular / Risks and Open Questions) one row at a time, proposing source-labeled answer candidates and merging the chosen answer into the proper body section. The pattern follows a spec-resolver approach (walk the open rows, propose source-labeled answer candidates, merge the chosen answer) adapted to the analysis doc contract.
 
 **Core invariant - read this twice:** the analysis doc is authoritative and forward-looking. The resolver never invents an answer; if no source produces a credible candidate, the only options offered are Defer and Other. Each Section 20 row is its own decision; never blend candidates across rows.
 
-> **Language**: Per `refs/rules.md` Language Application matrix - instruction prose stays English. `AskUserQuestion.label` and `header` stay English; `question` and `description` follow `prefs.global.outputLanguage`. Fragments written INTO the doc follow the doc's own front-matter `language` field (not `outputLanguage`), so a `tr` doc stays uniformly Turkish even when the session language differs.
+> **Language**: Per `$HOME/.claude/multi-agent-refs/rules.md` Language Application matrix - instruction prose stays English. `AskUserQuestion.label` and `header` stay English; `question` and `description` follow `prefs.global.outputLanguage`. Fragments written INTO the doc follow the doc's own front-matter `language` field (not `outputLanguage`), so a `tr` doc stays uniformly Turkish even when the session language differs.
 
 ## Inherited Locked decisions (from /multi-agent:analysis - binding here)
 
@@ -117,8 +117,8 @@ For each open row in source order:
 
 | Path | Reason |
 |------|--------|
-| `refs/analysis-template.md` | Section 20 / Section 23 table contracts, front-matter shape, omission re-flow |
-| `refs/conventions-defaults.md` | `convention-fallback` default values per platform |
+| `$HOME/.claude/multi-agent-refs/analysis-template.md` | Section 20 / Section 23 table contracts, front-matter shape, omission re-flow |
+| `$HOME/.claude/multi-agent-refs/conventions-defaults.md` | `convention-fallback` default values per platform |
 | `~/.claude/lib/extract-conventions.sh` | fresh single-field extraction for convention-fallback candidates |
 | `pipeline/skills/shared/external/humanizer/SKILL.md` | tone reference for longer merged fragments (short fragments only need the punctuation gate) |
 

package/pipeline/commands/multi-agent/{autopilot.md → autopilot/SKILL.md}

@@ -7,7 +7,7 @@ argument-hint: '"task"  -  issue URL, Jira ID, free-text, or #id (for resume)'
 
 **Input**: $ARGUMENTS
 
-> **Language (read FIRST)**: Before any status output, read `prefs.global.outputLanguage` and render every conversational line in it. `AskUserQuestion` and external payloads stay English. Full contract: `refs/rules.md` "Language Application".
+> **Language (read FIRST)**: Before any status output, read `prefs.global.outputLanguage` and render every conversational line in it. `AskUserQuestion` and external payloads stay English. Full contract: `$HOME/.claude/multi-agent-refs/rules.md` "Language Application".
 
 Run the task end-to-end with no confirmations.
 
@@ -50,7 +50,7 @@ Run the task end-to-end with no confirmations.
 4. **On error**  -  after 3 failed retries, pause and ask the user
 ## Required: Phase Tracker Contract
 
-**The phase tracker is mandatory**  -  the agent cannot skip it. Full spec: [`refs/tracker-contract.md`](./refs/tracker-contract.md).
+**The phase tracker is mandatory**  -  the agent cannot skip it. Full spec: [`$HOME/.claude/multi-agent-refs/tracker-contract.md`]($HOME/.claude/multi-agent-refs/tracker-contract.md).
 
 > **Autopilot mode:** user confirmations are skipped. The tracker is still mandatory  -  autopilot agent calls cannot skip it; skipping breaks `smoke-tracker-contract.sh`.
 
@@ -78,7 +78,7 @@ bash $HOME/.claude/scripts/phase-tracker.sh tokens <N> <in> <out>
 
 In Claude Code the agent MUST also drive the native TaskList widget so the user sees a sticky phase tile stack  -  this is the only progress signal Claude Code surfaces. Skipping these calls is the #1 source of "I don't see any phases" complaints.
 
-**TaskCreate ordering (strict)**: All TaskCreate calls fire in strict phase-number order BEFORE any TaskUpdate is applied. The native widget renders by creation order, not by phase number  -  out-of-order calls produce visually scrambled tile stacks (e.g. `1 ✓ · 2 ✓ · 4 ✓ · 0 ▶ · 3 ☐`) even when the underlying state is correct. Pre-marking phases as completed/skipped before Phase 0 starts is FORBIDDEN  -  register the tile in order, then flip status via TaskUpdate when the phase actually short-circuits. Full contract in `refs/tracker-contract.md` section "TaskCreate ordering (strict)".
+**TaskCreate ordering (strict)**: All TaskCreate calls fire in strict phase-number order BEFORE any TaskUpdate is applied. The native widget renders by creation order, not by phase number  -  out-of-order calls produce visually scrambled tile stacks (e.g. `1 ✓ · 2 ✓ · 4 ✓ · 0 ▶ · 3 ☐`) even when the underlying state is correct. Pre-marking phases as completed/skipped before Phase 0 starts is FORBIDDEN  -  register the tile in order, then flip status via TaskUpdate when the phase actually short-circuits. Full contract in `$HOME/.claude/multi-agent-refs/tracker-contract.md` section "TaskCreate ordering (strict)".
 
 ```text
 # Phase 0 startup  -  register one tile per phase (0..N), capture the taskId, persist it:
@@ -103,7 +103,7 @@ bash $HOME/.claude/scripts/phase-tracker.sh update <N> completed
 
 #### TaskCreate ordering (strict)
 
-**All TaskCreate calls fire in strict phase-number order BEFORE any TaskUpdate is applied.** For `autopilot` that means: Phase 0 → Phase 1 → Phase 2 → Phase 3 → Phase 4 → Phase 6 → Phase 7. The native widget renders by creation order, not by phase number  -  out-of-order calls produce visually scrambled tile stacks. Full ordering contract in `refs/tracker-contract.md` section "TaskCreate ordering (strict)".
+**All TaskCreate calls fire in strict phase-number order BEFORE any TaskUpdate is applied.** For `autopilot` that means: Phase 0 → Phase 1 → Phase 2 → Phase 3 → Phase 4 → Phase 6 → Phase 7. The native widget renders by creation order, not by phase number  -  out-of-order calls produce visually scrambled tile stacks. Full ordering contract in `$HOME/.claude/multi-agent-refs/tracker-contract.md` section "TaskCreate ordering (strict)".
 
 ### Visual channel  -  Copilot CLI / plain shell
 

package/pipeline/commands/multi-agent/{build-optimize.md → build-optimize/SKILL.md}

@@ -7,7 +7,7 @@ argument-hint: "(none - operates on current repo)"
 
 Wraps the vendored `xcode-build-orchestrator` skill (and its 5 specialist analyzers) so iOS users do not have to remember the orchestrator skill name. Recommend-first: no project files are modified without explicit developer approval. All output lands in `.build-benchmark/` in the current repo.
 
-> **Language**: Per `refs/rules.md` Language Application matrix - instruction prose stays English (this file is read as a system prompt). `AskUserQuestion.label` and `header` stay English. `question` and `description` follow `outputLanguage`. The emitted benchmark plan body (the upstream skill controls it) is English by upstream convention.
+> **Language**: Per `$HOME/.claude/multi-agent-refs/rules.md` Language Application matrix - instruction prose stays English (this file is read as a system prompt). `AskUserQuestion.label` and `header` stay English. `question` and `description` follow `outputLanguage`. The emitted benchmark plan body (the upstream skill controls it) is English by upstream convention.
 
 ## When to use
 

package/pipeline/commands/multi-agent/{channels.md → channels/SKILL.md}

@@ -320,7 +320,7 @@ Skipped in exactly two cases:
 {body}
 ```
 
-Secondary artifacts ride along as one-line intents at the bottom of the preview: Wiki page list, wiki → Jira triad comment (when Wiki is selected and the triad is active per `refs/issue-jira-triad.md`), Board status move (`{from} → {to}`), GitHub issue comment + Progress-flag sync (Phase 7 Step 1.5  -  same gate, see `refs/phases/phase-7-report.md`).
+Secondary artifacts ride along as one-line intents at the bottom of the preview: Wiki page list, wiki → Jira triad comment (when Wiki is selected and the triad is active per `$HOME/.claude/multi-agent-refs/issue-jira-triad.md`), Board status move (`{from} → {to}`), GitHub issue comment + Progress-flag sync (Phase 7 Step 1.5  -  same gate, see `$HOME/.claude/multi-agent-refs/phases/phase-7-report.md`).
 
 **Then a single AskUserQuestion** (`question`/`description` in `outputLanguage`; `label`/`header` English):
 
@@ -354,17 +354,17 @@ For each selected **channel**, call the corresponding adapter. Adapters run **in
 #### Adapter: PR description
 Aggregated Markdown from Step 5 → PR description. GitHub uses `gh pr edit --body-file`; Bitbucket uses a **reviewer-preserving (required)** `PUT /pull-requests/{id}` payload (must re-send `reviewers`, `fromRef`, `toRef`, `draft`, `version`  -  omitting any one resets that field server-side). Default is **replace**; `--append` opt-in. Multi-repo cross-links handled by `channels-multi-repo.sh render-pr`.
 
-Full contract: [`refs/channels/pr.md`](refs/channels/pr.md)  -  Bitbucket payload assembly snippet, version-mismatch retry, `--ready` promotion, multi-repo cross-link block.
+Full contract: [`$HOME/.claude/multi-agent-refs/channels/pr.md`]($HOME/.claude/multi-agent-refs/channels/pr.md)  -  Bitbucket payload assembly snippet, version-mismatch retry, `--ready` promotion, multi-repo cross-link block.
 
 #### Adapter: Jira comment
 Body converted to Jira wiki markup (`### ...` → `*...*`, `- [ ] ...` → `# ...`, `` `x` `` → `{{x}}`, tables to `||h||h|| |c|c|`); first line is the PR URL. POST `/rest/api/2/issue/{id}/comment` via heredoc + `jq --rawfile` + `curl --data-binary @file`. Token resolved from `keychainMapping.jira`.
 
-Full contract: [`refs/channels/jira.md`](refs/channels/jira.md)  -  full conversion table, multi-repo PR-list prepend, Wiki→Jira triad interaction.
+Full contract: [`$HOME/.claude/multi-agent-refs/channels/jira.md`]($HOME/.claude/multi-agent-refs/channels/jira.md)  -  full conversion table, multi-repo PR-list prepend, Wiki→Jira triad interaction.
 
 #### Adapter: Confluence page
 Token from `keychainMapping.confluence`; missing/expired triggers the inline Save Flow from `setup.md`. Parent page picked from `prefs.projects[<project>].confluenceUrls` recents (LRU, max 5) or paste new URL. Auto-suggest title `"{jiraId}  -  {taskTitle}"`. Body converted to Confluence storage format (`<h3>`, `<table>`, `<ac:structured-macro>` code blocks). POST `/rest/api/content`; URL saved back to recents.
 
-Full contract: [`refs/channels/confluence.md`](refs/channels/confluence.md)  -  URL parsing rules, full storage-XML conversion table, update-vs-create branch.
+Full contract: [`$HOME/.claude/multi-agent-refs/channels/confluence.md`]($HOME/.claude/multi-agent-refs/channels/confluence.md)  -  URL parsing rules, full storage-XML conversion table, update-vs-create branch.
 
 #### Adapter: Board (GitHub Projects v2 status move)
 
@@ -385,13 +385,13 @@ Full contract: [`refs/channels/confluence.md`](refs/channels/confluence.md)  -
 **Additional behaviors handled inline by this adapter** (no separate contract file): token scope upgrade flow (`gh auth refresh -s project,read:project`), field-ID auto-discovery for repos missing `projectV2Fields.status` in figma-config, error-recovery (token expired vs scope insufficient).
 
 #### Adapter: Wiki
-**Precondition gate:** `state.taskType === "component"` AND `figmaConfig.wiki.enabled === true`. If not met, shows the Case B menu from `refs/wiki-capture.md` (setup / skip / disable / manual-note).
+**Precondition gate:** `state.taskType === "component"` AND `figmaConfig.wiki.enabled === true`. If not met, shows the Case B menu from `$HOME/.claude/multi-agent-refs/wiki-capture.md` (setup / skip / disable / manual-note).
 
-**Wiki → Jira comment auto-link** (triad contract, preserved from v5.6): when the Wiki adapter writes pages AND `state.jiraId` is set AND `prefs.global.wikiToJiraComment === true`, also post a humanizer-passed summary of the wiki pages (component name + variant count + wiki URL) as an additional Jira comment on the same issue. Independent of the main Jira channel's comment  -  both can coexist. Full contract in `refs/issue-jira-triad.md`.
+**Wiki → Jira comment auto-link** (triad contract, preserved from v5.6): when the Wiki adapter writes pages AND `state.jiraId` is set AND `prefs.global.wikiToJiraComment === true`, also post a humanizer-passed summary of the wiki pages (component name + variant count + wiki URL) as an additional Jira comment on the same issue. Independent of the main Jira channel's comment  -  both can coexist. Full contract in `$HOME/.claude/multi-agent-refs/issue-jira-triad.md`.
 
-**If preconditions met (Case A):** scope multi-select prompt (Main page / iOS sub-page / Screenshots / Index updates / Other); adapter dispatched to `pipeline/skills/figma-ios/figma-component-wiki` (or `figma-android/`).
+**If preconditions met (Case A):** scope multi-select prompt (Main page / iOS sub-page / Screenshots / Index updates / Other); adapter dispatched to the plugin skill `ai-ios-engineering-toolkit:figma-component-wiki` (or `ai-android-engineering-toolkit:figma-component-wiki`).
 
-Full contract: [`refs/channels/wiki.md`](refs/channels/wiki.md)  -  all four `figma-config.wiki.mode` adapters (submodule / in-repo / github-wiki / separate-repo), full Case A scope menu, screenshot quadrant gate.
+Full contract: [`$HOME/.claude/multi-agent-refs/channels/wiki.md`]($HOME/.claude/multi-agent-refs/channels/wiki.md)  -  all four `figma-config.wiki.mode` adapters (submodule / in-repo / github-wiki / separate-repo), full Case A scope menu, screenshot quadrant gate.
 
 ### Step 8  -  Summary
 
@@ -446,7 +446,7 @@ Persist selections to `prefs.global.reportChannels` + `prefs.global.reportConten
 | No pipeline log + "Normal analiz" selected | Greyed out at menu; if forced via `--content`, error |
 | Bitbucket PR version mismatch | Refetch once, retry; fail on second mismatch |
 | Jira token missing/expired | Token Save Flow inline; skip on user-skip |
-| Wiki preconditions not met | Case B menu per `refs/wiki-capture.md` |
+| Wiki preconditions not met | Case B menu per `$HOME/.claude/multi-agent-refs/wiki-capture.md` |
 
 ## Phase 7 delegation contract
 
@@ -459,7 +459,7 @@ Phase 7 invokes this command internally instead of running its own inline Jira/C
 
 Phase 7 passes a runtime state bundle: `{jiraId, prUrl, remoteType, taskType, figmaConfig, agentState, pipelineLogPath}`. `channels` reads these directly, skips target-resolution Step 1.
 
-**Autopilot contract:** Phase 7 ALWAYS pauses at the channels menu (breaks generic autopilot zero-interaction). 30-min timeout → session ends cleanly, task resumable via `/multi-agent:resume`. See `refs/phases/modes.md` for full contract. The Step 6 body preview + approval gate follows the same split: interactive runs approve the composed bodies before dispatch; autopilot posts directly after the menu resolves.
+**Autopilot contract:** Phase 7 ALWAYS pauses at the channels menu (breaks generic autopilot zero-interaction). 30-min timeout → session ends cleanly, task resumable via `/multi-agent:resume`. See `$HOME/.claude/multi-agent-refs/phases/modes.md` for full contract. The Step 6 body preview + approval gate follows the same split: interactive runs approve the composed bodies before dispatch; autopilot posts directly after the menu resolves.
 
 ## Cross-CLI parity
 
@@ -468,8 +468,8 @@ Phase 7 passes a runtime state bundle: `{jiraId, prUrl, remoteType, taskType, fi
 | Claude Code | `/multi-agent:channels` | Slash namespace (colon) |
 | Copilot CLI | `multi-agent-channels` | Top-level command (dash-separated, no namespace support) |
 
-Both installations read the same source file (`pipeline/commands/multi-agent/channels.md`)  -  `install.js --all` copies to:
-- `~/.claude/commands/multi-agent/channels.md`
+Both installations read the same source file (`pipeline/commands/multi-agent/channels/SKILL.md`)  -  `install.js --all` copies to:
+- `~/.claude/commands/multi-agent/channels/SKILL.md`
 - `~/.copilot/commands/multi-agent-channels.md`
 
 Prompts MUST be byte-identical. `smoke-cross-cli-behavior.sh` asserts this  -  spacing, brackets, numbering, log-line format.

package/pipeline/commands/multi-agent/{dev.md → dev/SKILL.md}

@@ -18,7 +18,7 @@ description: "Fast development mode: Init → Dev (Opus) → Test → Commit →
 >   - Code identifiers, file paths, log lines
 >   - PR title prefix (`feat:`, `fix:`, ...)  -  only the body honors `outputLanguage`
 >
-> Full contract: `refs/rules.md` "Language Application".
+> Full contract: `$HOME/.claude/multi-agent-refs/rules.md` "Language Application".
 
 A 5-phase fast pipeline: Init → Dev → Test → Commit → Report. Development runs on the Opus model (top intelligence tier).
 
@@ -182,7 +182,7 @@ Phase 0 (Init full picker), Phase 5 (User Test), Phase 6 (Commit), and Phase 7 (
 | Duration | ~10-15 min | ~5-7 min |
 ## Required: Phase Tracker Contract
 
-**The phase tracker is mandatory**  -  the agent cannot skip it. Full spec: [`refs/tracker-contract.md`](./refs/tracker-contract.md).
+**The phase tracker is mandatory**  -  the agent cannot skip it. Full spec: [`$HOME/.claude/multi-agent-refs/tracker-contract.md`]($HOME/.claude/multi-agent-refs/tracker-contract.md).
 
 Two channels run in parallel at every phase boundary:
 
@@ -208,7 +208,7 @@ bash $HOME/.claude/scripts/phase-tracker.sh tokens <N> <in> <out>
 
 In Claude Code the agent MUST also drive the native TaskList widget so the user sees a sticky phase tile stack  -  this is the only progress signal Claude Code surfaces. Skipping these calls is the #1 source of "I don't see any phases" complaints.
 
-**TaskCreate ordering (strict)**: All TaskCreate calls fire in strict phase-number order BEFORE any TaskUpdate is applied. The native widget renders by creation order, not by phase number  -  out-of-order calls produce visually scrambled tile stacks (e.g. `1 ✓ · 2 ✓ · 4 ✓ · 0 ▶ · 3 ☐`) even when the underlying state is correct. Pre-marking phases as completed/skipped before Phase 0 starts is FORBIDDEN  -  register the tile in order, then flip status via TaskUpdate when the phase actually short-circuits. Full contract in `refs/tracker-contract.md` section "TaskCreate ordering (strict)".
+**TaskCreate ordering (strict)**: All TaskCreate calls fire in strict phase-number order BEFORE any TaskUpdate is applied. The native widget renders by creation order, not by phase number  -  out-of-order calls produce visually scrambled tile stacks (e.g. `1 ✓ · 2 ✓ · 4 ✓ · 0 ▶ · 3 ☐`) even when the underlying state is correct. Pre-marking phases as completed/skipped before Phase 0 starts is FORBIDDEN  -  register the tile in order, then flip status via TaskUpdate when the phase actually short-circuits. Full contract in `$HOME/.claude/multi-agent-refs/tracker-contract.md` section "TaskCreate ordering (strict)".
 
 ```text
 # Phase 0 startup  -  register one tile per phase (0..N), capture the taskId, persist it:
@@ -233,7 +233,7 @@ bash $HOME/.claude/scripts/phase-tracker.sh update <N> completed
 
 #### TaskCreate ordering (strict)
 
-**All TaskCreate calls fire in strict phase-number order BEFORE any TaskUpdate is applied.** For `--dev` that means: Phase 0 → Phase 3 → Phase 5 → Phase 6 → Phase 7. The native widget renders by creation order, not by phase number  -  out-of-order calls produce visually scrambled tile stacks. Full ordering contract in `refs/tracker-contract.md` section "TaskCreate ordering (strict)".
+**All TaskCreate calls fire in strict phase-number order BEFORE any TaskUpdate is applied.** For `--dev` that means: Phase 0 → Phase 3 → Phase 5 → Phase 6 → Phase 7. The native widget renders by creation order, not by phase number  -  out-of-order calls produce visually scrambled tile stacks. Full ordering contract in `$HOME/.claude/multi-agent-refs/tracker-contract.md` section "TaskCreate ordering (strict)".
 
 ### Visual channel  -  Copilot CLI / plain shell
 

package/pipeline/commands/multi-agent/{dev-autopilot.md → dev-autopilot/SKILL.md}

@@ -6,7 +6,7 @@ description: "Fastest mode: Dev (Opus) plus Autopilot. Init → Dev → Commit
 
 **Input**: $ARGUMENTS
 
-> **Language (read FIRST)**: Before any status output, read `prefs.global.outputLanguage` and render every conversational line in it. `AskUserQuestion` and external payloads stay English. Full contract: `refs/rules.md` "Language Application".
+> **Language (read FIRST)**: Before any status output, read `prefs.global.outputLanguage` and render every conversational line in it. `AskUserQuestion` and external payloads stay English. Full contract: `$HOME/.claude/multi-agent-refs/rules.md` "Language Application".
 
 Dev mode + Autopilot combined: a 4-phase pipeline with no confirmations, end-to-end autonomous.
 
@@ -54,7 +54,7 @@ Phase 7: Report     → short terminal summary
 | Estimated duration | ~12 min | ~5 min | ~10 min | **~3 min** |
 ## Required: Phase Tracker Contract
 
-**The phase tracker is mandatory**  -  the agent cannot skip it. Full spec: [`refs/tracker-contract.md`](./refs/tracker-contract.md).
+**The phase tracker is mandatory**  -  the agent cannot skip it. Full spec: [`$HOME/.claude/multi-agent-refs/tracker-contract.md`]($HOME/.claude/multi-agent-refs/tracker-contract.md).
 
 > **Autopilot mode:** user confirmations are skipped. The tracker is still mandatory  -  autopilot agent calls cannot skip it; skipping breaks `smoke-tracker-contract.sh`.
 
@@ -82,7 +82,7 @@ bash $HOME/.claude/scripts/phase-tracker.sh tokens <N> <in> <out>
 
 In Claude Code the agent MUST also drive the native TaskList widget so the user sees a sticky phase tile stack  -  this is the only progress signal Claude Code surfaces. Skipping these calls is the #1 source of "I don't see any phases" complaints.
 
-**TaskCreate ordering (strict)**: All TaskCreate calls fire in strict phase-number order BEFORE any TaskUpdate is applied. The native widget renders by creation order, not by phase number  -  out-of-order calls produce visually scrambled tile stacks (e.g. `1 ✓ · 2 ✓ · 4 ✓ · 0 ▶ · 3 ☐`) even when the underlying state is correct. Pre-marking phases as completed/skipped before Phase 0 starts is FORBIDDEN  -  register the tile in order, then flip status via TaskUpdate when the phase actually short-circuits. Full contract in `refs/tracker-contract.md` section "TaskCreate ordering (strict)".
+**TaskCreate ordering (strict)**: All TaskCreate calls fire in strict phase-number order BEFORE any TaskUpdate is applied. The native widget renders by creation order, not by phase number  -  out-of-order calls produce visually scrambled tile stacks (e.g. `1 ✓ · 2 ✓ · 4 ✓ · 0 ▶ · 3 ☐`) even when the underlying state is correct. Pre-marking phases as completed/skipped before Phase 0 starts is FORBIDDEN  -  register the tile in order, then flip status via TaskUpdate when the phase actually short-circuits. Full contract in `$HOME/.claude/multi-agent-refs/tracker-contract.md` section "TaskCreate ordering (strict)".
 
 ```text
 # Phase 0 startup  -  register one tile per phase (0..N), capture the taskId, persist it:
@@ -107,7 +107,7 @@ bash $HOME/.claude/scripts/phase-tracker.sh update <N> completed
 
 #### TaskCreate ordering (strict)
 
-**All TaskCreate calls fire in strict phase-number order BEFORE any TaskUpdate is applied.** For `--dev autopilot` that means: Phase 0 → Phase 3 → Phase 6 → Phase 7. The native widget renders by creation order, not by phase number  -  out-of-order calls produce visually scrambled tile stacks. Full ordering contract in `refs/tracker-contract.md` section "TaskCreate ordering (strict)".
+**All TaskCreate calls fire in strict phase-number order BEFORE any TaskUpdate is applied.** For `--dev autopilot` that means: Phase 0 → Phase 3 → Phase 6 → Phase 7. The native widget renders by creation order, not by phase number  -  out-of-order calls produce visually scrambled tile stacks. Full ordering contract in `$HOME/.claude/multi-agent-refs/tracker-contract.md` section "TaskCreate ordering (strict)".
 
 ### Visual channel  -  Copilot CLI / plain shell