@mmerterden/multi-agent-pipeline 20.23.1 → 20.24.0

package/CHANGELOG.md

@@ -16,11 +16,40 @@ Internal file-layout changes that don't affect the slash-command surface are sti
 
 ## [Unreleased]
 
+## [20.24.0] - 2026-10-14
+
+### Added
+
+- **Plan checks before approval.** `validate-planning.mjs` pairs every two tasks that touch the same file or name the same symbol in the optional `creates` / `removes` / `produces` / `consumes` lists and returns one anchored row per pair: two creators of one file fail the plan; removing what another task uses, a consumer not ordered after its producer and an unordered shared edit are warnings shown at the approval gate. `plan-complexity.mjs` scores each task 1-10 from the repository (files listed, code-graph fan-in of their declarations, modules spanned, dependents outside them, test surface); a task above `planComplexity.splitAbove` (default 7) is split before approval with its requirement ids carried onto the parts. Contract: `multi-agent-refs/features/plan-checks.md`.
+- **Drift holds what was built on it.** `plan-todos.sh drift` compares the files a step changed with its planned `files[]`, `creates[]` and their test files (paths normalised) and takes its not-done count; every pending step that depends on it, directly or transitively, gets `needsReplan`, `next` reports it as waiting, `start` and `complete` refuse it, and `confirm-replan` releases it once its plan has been re-read. Applies where Phase 2 iterates plan todos (`planTodos.enabled`).
+- **TDD evidence per todo.** `plan-todos.sh tdd-red` / `tdd-green` read the runner's own results (JUnit XML, an `.xcresult` bundle or its exported summary, xcodebuild / `swift test` logs) through `lib/tdd-evidence.mjs` and record `plan.todos[].tdd`; `complete` checks it per `tddEvidence.mode` (`off`, `warn` by default, `enforce`). Docs and chore todos, component tasks, test policy `none` and `complete --no-tests <reason>` are exempt with the reason recorded, and Phase 3 gives the reviewers one note per todo completed `unverified` or exempted by `--no-tests`. It applies where Phase 2 iterates plan todos (`planTodos.enabled`).
+- **Rework escalation.** On the last rework round the cap allows, or when `review-delta` reports findings still present after a fix, `rework-escalation.mjs` dispatches a fresh dev agent one rung up the model-fallback ladder with the triage round and the prior attempt's diff as files. It never escalates against the cost budget, on a run that fell back, onto fable while that rung is off, or past `reworkEscalation.maxTier` (default `opus`), and every refusal is recorded with its reason.
+- **Decisions made on your behalf.** Autopilot and unattended runs record every choice a person would otherwise have made in `state.decisions[]` as `{what, why, costIfWrong, at, phase}`: parks, blockers lowered to important, rework escalations, pickers resolved without asking, split consensus, date-gate fallbacks. The Phase 5 report and the autopilot PR body list them in one redacted section, and the PR body still passes the outbound gate.
+- **Mutation probe on changed lines** (opt-in, `mutationProbe.enabled`). `mutation-probe.mjs` makes up to `mutationProbe.maxMutants` (5) simple mutants on the lines the diff added or changed, runs the test command in a disposable detached worktree, and reports each surviving mutant as an `important` finding that the tests are weak. The user's checkout is never written. Phase 3 Step 1.763; recipe in `features/mutation-probe.md`.
+- **Syntax check after an edit** (opt-in, Claude Code). Set `editSyntaxGate.enabled` and merge the hooks block again through `/multi-agent:setup` (Step 8), which now carries the `PostToolUse` registration. The hook parses the edited file within 2 s (`swiftc -parse`, `node --check`, `py_compile` into a temp dir, JSON, `bash -n`) and tells the model about a syntax error; it never blocks, and stays silent when the parser is missing or slow.
+- **`find-polluter.mjs`** bisects a test list from a clean state to the test that leaves a stray file (`--absent`) or breaks another test only when run before it (`--target`), and stops with exit 2 when state survives a trial and no `--reset` was given; `debugging-strategies` points at it.
+- **`/multi-agent:postmortem [#id|transcript.jsonl] [--audit]`**: a read-only post-mortem of a run from its session transcripts. `transcript-index.mjs` indexes turns, loops of identical tool calls, stalls, compactions, subagents, errors and denials, token use and invocations, each cited as `file:line`, and redacts every excerpt (credentials, emails, hosts, paths, home directories) with a count. The report goes to the run's log directory; nothing is sent. Claude Code transcripts only.
+- **Story dependencies in `/multi-agent:analysis-jira`.** A requirement row's `Depends on: <ids>` clause links the stories in Jira (`issueTree.dependencyLinkType`, default `Blocks`, prerequisite to dependent); a dependency cycle or a prerequisite no story covers makes the plan `incomplete`. Documents without the clause plan exactly as before.
+- **Reply shape.** `rules/response-shape.md` (with its detail in `multi-agent-refs/response-shape.md`) shapes conversational replies: the answer or next action first, numbered single-action steps, at most five items shown without ever dropping a finding, errors as cause then fix, and a last line that says what happens next.
+- **Maintainer evals** (not shipped; their live calls never run in `npm test`, which tests them against a fake CLI): `eval-skill-compliance-live.mjs` pressure-tests discipline rules against a no-rule baseline with deterministic graders, and `eval-ab-judge.mjs` judges prompt and style variants blind, with hash-stable labels, a judge-only rubric region and a blocker-first verdict. Both are dry-run by default and need `--live`, `MULTI_AGENT_LIVE_EVAL=1` and an explicit budget; documented in `pipeline/eval/README.md`.
+- The App Store optimization skill orders the brief for a marketing frame around a real capture: device and size, the capture as the asset, verbatim captions per locale, brand tokens, explicit exclusions.
+
+### Changed
+
+- `plan-todos.sh next` (with `planTodos.enabled`) answers from a task graph: the ready todo that unblocks the most pending work comes first, then the longest chain behind it, then plan order, and a todo a failed or missing dependency or a cycle blocks is named on stderr instead of leaving `next` silently empty.
+
+## [20.23.2] - 2026-10-13
+
+### Added
+
+- `issueTree.storyPanels` sets the colours of a Story's wiki panels (`story`, `criteria`, `other`, each `#rrggbb`) or turns them off with `false`; `/multi-agent:create-jira` and `/multi-agent:analysis-jira` both read it. Unset keeps `#d4d39e`, `#d4a19e` and `#9eb6d4`, and a value that is not a colour keeps its default.
+- `/multi-agent:analysis-jira` asks once, in an attended run with no `issueTree.subtaskRoles`, which sub-tasks every story gets, offering the site's own sub-task types (`analysis-jira-write.sh --subtask-types`), and saves the answer (`analysis-story-tree.mjs --save-subtasks`); "no sub-tasks" is remembered as `issueTree.subtaskRolesDeclined`.
+
 ## [20.23.1] - 2026-10-13
 
 ### Fixed
 
-- `/multi-agent:analysis-jira` finds a site's sub-task types on Jira Data Center 9, which answers only the per-project `createmeta/{project}/issuetypes`; sub-tasks were created under the generic `Sub-task` name there. `issueTree.subtaskIssueType` may now map each role to the site's type (`{"develop": "Development Task", "test": "Test Task"}`, `"*"` for any other role), a site with several sub-task types and no setting is named in the run's output, and the preview shows the type each role gets.
+- `/multi-agent:analysis-jira` finds a site's sub-task types on Jira Data Center 9, which answers only the per-project `createmeta/{project}/issuetypes`; sub-tasks were created under the generic `Sub-task` name there. `issueTree.subtaskIssueType` may now map each role to the site's type (`{"develop": "Build Sub-task", "test": "QA Sub-task"}`, `"*"` for any other role), a site with several sub-task types and no setting is named in the run's output, and the preview shows the type each role gets.
 - The code graph reads Swift imports behind an attribute (`@testable import`, `@_exported import`, `@preconcurrency import`, `@_spi(...) import`).
 - Python type parameters (`class Box[T](Base):`) keep their bases, and a TypeScript base built by a call (`extends mixin(Base)`) adds no edge to the function.
 - A diff hunk that only inserts or only deletes lines is given to the declaration spanning both lines around it, at any context size; at `-U0` it was given to the line before it.

package/README.md

@@ -17,7 +17,7 @@ Runs natively on Claude Code, Copilot CLI and Codex CLI. macOS only. Zero runtim
 ### Prerequisites
 
 - **Node.js >= 20.11** - required; the pipeline's own tooling runs on it.
-- **`jq`** - required for nine paths, optional for the rest. 92 shell files call it. The nine that publish or decide - the autopilot queue, Jira comments, PR reviews, issue updates, the plan file, both Figma fetchers, log search and Jira auth - refuse with exit 3 rather than run, because a missing `jq` renders as empty DATA and the work carries on with it. Everywhere else it still degrades. The install prints a note when it is missing.
+- **`jq`** - required for nine paths, optional for the rest. 93 shell files call it. The nine that publish or decide - the autopilot queue, Jira comments, PR reviews, issue updates, the plan file, both Figma fetchers, log search and Jira auth - refuse with exit 3 rather than run, because a missing `jq` renders as empty DATA and the work carries on with it. Everywhere else it still degrades. The install prints a note when it is missing.
 - **`gh`** - for GitHub issue and PR work. Its built-in `--jq` is independent of the `jq` binary.
 
 ## Quick Start
@@ -158,7 +158,7 @@ The discipline behind all of this - bounded loops, evidence gates, token-budgete
 
 ## Commands
 
-`/multi-agent` plus 64 sub-commands. `/multi-agent:help` renders the same catalog in your terminal, in your `outputLanguage`.
+`/multi-agent` plus 65 sub-commands. `/multi-agent:help` renders the same catalog in your terminal, in your `outputLanguage`.
 
 ### Pipeline entries
 
@@ -192,6 +192,7 @@ The discipline behind all of this - bounded loops, evidence gates, token-budgete
 | `/multi-agent:review-analysis`     | Review a written analysis document; findings cite the Locked rule they break                |
 | `/multi-agent:diff-explain`        | Map a Phase 3 triage finding back to the diff lines that caused it                          |
 | `/multi-agent:estimate`            | Effort range from similar closed work: P25-P75 per role, with the tickets behind it         |
+| `/multi-agent:postmortem`          | Post-mortem of a run from its transcripts: loops, stalls, root-cause hypotheses on file:line |
 | `/multi-agent:scenario-audit`      | Check test scenarios against the code; each PRESENT on a resolving file:line                |
 | `/multi-agent:refactor`            | Best-practice extraction + bug hunt + derived-skill drift + toolkit MCP research → one plan |
 | `/multi-agent:scan`                | Skill security scan of local skill directories against a tiered pattern catalog             |
@@ -406,17 +407,17 @@ This enables the matching plugin (+ the shared `ai-common` plugin) in the repo's
 
 ## Tool support
 
-The pipeline runs natively on **Claude Code**, **Copilot CLI** and **Codex CLI** - all three install from the same `pipeline/` source and get the same 64 commands.
+The pipeline runs natively on **Claude Code**, **Copilot CLI** and **Codex CLI** - all three install from the same `pipeline/` source and get the same 65 commands.
 
 | Tool        | Flag                 | What it installs                                                                                       |
 | ----------- | -------------------- | ------------------------------------------------------------------------------------------------------ |
 | Claude Code | `--claude` (default) | slash commands + skills + agents + three `PreToolUse` hooks (secret scan, agent-guard, read-size gate) |
-| Copilot CLI | `--copilot`          | instructions + 64 sub-command skills + scripts                                                         |
-| Codex CLI   | `--codex`            | one router skill + 64 specs as refs + 10 agent TOML + `AGENTS.md` block + `codex mcp add`               |
+| Copilot CLI | `--copilot`          | instructions + 65 sub-command skills + scripts                                                         |
+| Codex CLI   | `--codex`            | one router skill + 65 specs as refs + 10 agent TOML + `AGENTS.md` block + `codex mcp add`               |
 
 Filter skills by stack with `--platform=ios\|android\|all`.
 
-**Why Codex gets one skill and not 64.** Codex assembles every discovered skill's name
+**Why Codex gets one skill and not 65.** Codex assembles every discovered skill's name
 and description into a single prompt block and drops entries when it overflows, with no
 error. Measured on 0.145: installing one plugin that declares 142 skills surfaced only
 75 of them and evicted an unrelated user skill. So on Codex the pipeline ships a single

package/README.tr.md

@@ -157,7 +157,7 @@ Koşunun kendisinde ayarlanabilen tek şey autopilot; geri kalan her şey kendi
 
 ## Komutlar
 
-`/multi-agent` ve 64 alt komut. `/multi-agent:help` aynı katalogu terminalde, `outputLanguage` ayarına göre gösterir.
+`/multi-agent` ve 65 alt komut. `/multi-agent:help` aynı katalogu terminalde, `outputLanguage` ayarına göre gösterir.
 
 ### Pipeline girişleri
 
@@ -191,6 +191,7 @@ Koşunun kendisinde ayarlanabilen tek şey autopilot; geri kalan her şey kendi
 | `/multi-agent:review-analysis`     | Yazılmış analiz dokümanını review eder; bulgular ihlal edilen Locked kuralını gösterir           |
 | `/multi-agent:diff-explain`        | Faz 4 triyaj bulgusunu onu doğuran diff satırlarına eşler                                        |
 | `/multi-agent:estimate`            | Benzer kapanmış işlerden efor aralığı: rol başına P25-P75, dayandığı ticket'larla               |
+| `/multi-agent:postmortem`          | Koşunun transkriptlerinden post-mortem: döngüler, takılmalar, dosya:satır dayanaklı kök-neden hipotezleri |
 | `/multi-agent:scenario-audit`      | Test senaryolarını koda karşı denetler; her VAR çözülen bir dosya:satıra dayanır                 |
 | `/multi-agent:refactor`            | Best-practice çıkarımı + bug avı + türetilmiş skill drift'i + toolkit MCP araştırması → tek plan |
 | `/multi-agent:scan`                | Yerel skill dizinlerini kademeli desen kataloğuna göre güvenlik taraması                         |
@@ -405,17 +406,17 @@ Bu, ilgili plugin'i (+ ortak `ai-common` plugin'ini) repo'nun `.claude/settings.
 
 ## Araç desteği
 
-Pipeline **Claude Code**, **Copilot CLI** ve **Codex CLI** üzerinde native çalışır - üçü de aynı `pipeline/` kaynağından kurulur ve aynı 64 komutu alır.
+Pipeline **Claude Code**, **Copilot CLI** ve **Codex CLI** üzerinde native çalışır - üçü de aynı `pipeline/` kaynağından kurulur ve aynı 65 komutu alır.
 
 | Araç        | Bayrak                  | Ne kurar                                                                                                         |
 | ----------- | ----------------------- | ---------------------------------------------------------------------------------------------------------------- |
 | Claude Code | `--claude` (varsayılan) | slash komutları + skill'ler + agent'lar + üç `PreToolUse` hook'u (secret scan, agent-guard, okuma-boyutu geçidi) |
-| Copilot CLI | `--copilot`             | talimatlar + 64 alt-komut skill'i + script'ler                                                                   |
-| Codex CLI   | `--codex`               | bir router skill + ref olarak 64 spec + 10 agent TOML + `AGENTS.md` bloğu + `codex mcp add`                       |
+| Copilot CLI | `--copilot`             | talimatlar + 65 alt-komut skill'i + script'ler                                                                   |
+| Codex CLI   | `--codex`               | bir router skill + ref olarak 65 spec + 10 agent TOML + `AGENTS.md` bloğu + `codex mcp add`                       |
 
 Skill'leri stack'e göre filtrele: `--platform=ios\|android\|all`.
 
-**Codex neden 64 değil de tek bir skill alıyor.** Codex, keşfettiği her skill'in adını
+**Codex neden 65 değil de tek bir skill alıyor.** Codex, keşfettiği her skill'in adını
 ve açıklamasını tek bir prompt bloğuna toplar ve blok taştığında girdileri hatasızca
 düşürür. 0.145 üzerinde ölçüldü: 142 skill deklare eden bir plugin kurulduğunda sadece
 75'i yüzeye çıktı ve alakasız bir kullanıcı skill'i tahliye edildi. Bu yüzden Codex'te

package/docs/architecture.md

@@ -113,7 +113,7 @@ graph TB
     end
 
     subgraph "Pipeline Specs"
-        CMD[commands/<br/>64 command files]
+        CMD[commands/<br/>65 command files]
         AGT[agents/<br/>10 agent personas]
         RUL[rules/<br/>13 domain rules]
         PHS[multi-agent-refs/phases/<br/>phase specs + contracts]
@@ -204,8 +204,8 @@ repo and the website, plus the two independently shipped repos
 ```mermaid
 graph TD
     CC["Claude Code<br/>(source of truth)"]
-    COP["Copilot CLI<br/>(instructions + 64 sub-command skills)"]
-    COD["Codex CLI<br/>(1 router skill + 64 refs)"]
+    COP["Copilot CLI<br/>(instructions + 65 sub-command skills)"]
+    COD["Codex CLI<br/>(1 router skill + 65 refs)"]
     REPO["Pipeline Repo<br/>(npm package)"]
     WEB["Website"]
     PLUGREPO["multi-agent-plugins<br/>(6 plugins, own repo)"]

package/docs/ecosystem.md

@@ -5,7 +5,7 @@ separately, wired together at install time and at run time:
 
 | Repo | What it owns | Ships as |
 |---|---|---|
-| **`multi-agent-pipeline`** (this repo) | Orchestration: the 6-phase flow, the 64 slash commands, quality gates, review/triage, cross-CLI parity | npm package (`@mmerterden/multi-agent-pipeline`), installs itself onto Claude Code / Copilot CLI / Codex CLI |
+| **`multi-agent-pipeline`** (this repo) | Orchestration: the 6-phase flow, the 65 slash commands, quality gates, review/triage, cross-CLI parity | npm package (`@mmerterden/multi-agent-pipeline`), installs itself onto Claude Code / Copilot CLI / Codex CLI |
 | **`multi-agent-plugins`** | Stack knowledge: per-platform component/lifecycle skills (iOS, Android, Frontend, Backend) + shared knowledge | Claude Code marketplace, 6 independently-versioned plugins |
 | **`multi-agent-toolkit-mcp`** | The pipeline's hands on devices and browsers: 197 MCP tools across 21 categories (simulator/emulator control, end-to-end locators and assertions, memory, crash diagnostics, accessibility audit, store compliance, web automation and scraping, Figma-vs-mock design audit, code intelligence, wallet passes, an agent-DSL batch runner, a full-text context index with text-to-image rendering, provider-backed research, verified hand-off of simple text tasks to free models, video key frames, offline security scoring, read-only access to related repositories at their latest branch, travel document fixtures, a voice channel, App Store Connect and Play Console APIs, virtual iPhones) | npm package, registered as a standard stdio MCP server on every host |
 
@@ -18,7 +18,7 @@ Either can be swapped or removed without touching the other two's source.
 graph LR
     subgraph PIPE ["multi-agent-pipeline (orchestrator)"]
         direction TB
-        PHASES["6 phases · 64 commands"]
+        PHASES["6 phases · 65 commands"]
         GATES["deterministic gates + review triage"]
     end
 
@@ -68,8 +68,8 @@ only those:
 graph TD
     CC["Claude Code<br/>~/.claude/commands/multi-agent/<br/>(source of truth)"]
 
-    CC -->|"Step 2: copy + reformat<br/>64 sub-command skills"| COP["Copilot CLI<br/>~/.copilot/skills/"]
-    CC -->|"Step 2b: transform<br/>(install.js --codex)"| COD["Codex CLI<br/>1 router skill + 64 refs<br/>+ 10 agent TOML"]
+    CC -->|"Step 2: copy + reformat<br/>65 sub-command skills"| COP["Copilot CLI<br/>~/.copilot/skills/"]
+    CC -->|"Step 2b: transform<br/>(install.js --codex)"| COD["Codex CLI<br/>1 router skill + 65 refs<br/>+ 10 agent TOML"]
     CC -->|"Step 3: genericize<br/>(strip personal data)"| REPO["multi-agent-pipeline repo<br/>pipeline/"]
     CC -->|"Step 4: version + feature sync"| WEB["Website<br/>projects.ts / i18n.tsx"]
 
@@ -161,7 +161,7 @@ measurements behind this table):
 
 | | Claude Code | Copilot CLI | Codex CLI |
 |---|---|---|---|
-| **Pipeline commands** | 64 slash-command skills, native | 64 sub-command skills, `multi-agent-{cmd}` naming, copied in | 1 router skill (`multi-agent`) + 64 command specs as reference files - Codex silently truncates its skills block past a few dozen entries, so sub-commands are not peer skills here |
+| **Pipeline commands** | 65 slash-command skills, native | 65 sub-command skills, `multi-agent-{cmd}` naming, copied in | 1 router skill (`multi-agent`) + 65 command specs as reference files - Codex silently truncates its skills block past a few dozen entries, so sub-commands are not peer skills here |
 | **Stack plugins** | Marketplace plugin, loaded natively, resolved by `.claude/settings.json` enabled-list | Enabled plugin's authored skills copied flat into `~/.copilot/skills/`; `knowledge/` **not** re-copied (already delivered via `shared/external`) | Copied as reference files under `~/.codex/multi-agent-refs/skills/`, plugin-prefixed on name clash (e.g. `architecture` → `ai-ios-toolkit-architecture`) |
 | **Component dispatch (Phase 3)** | Marketplace plugin's `create-component`/`create-screen` skill via the Skill tool | No plugin loader - the enabled stack plugin's authored skills (incl. `create-component`) are copied flat into `~/.copilot/skills/` at install time (the old frozen `figma-*` copies are pruned, they were never a fallback) | Not part of the enforced parity axis; classification + state-shape must match, skill *inventory* does not |
 | **multi-agent-toolkit-mcp** | `claude mcp add multi-agent-toolkit -- npx -y @mmerterden/multi-agent-toolkit-mcp` | `copilot mcp add multi-agent-toolkit -- npx -y @mmerterden/multi-agent-toolkit-mcp` | `codex mcp add multi-agent-toolkit -- npx -y @mmerterden/multi-agent-toolkit-mcp` (skipped with a warning if `codex` isn't on `PATH`) |

package/docs/facts.json

@@ -1,7 +1,7 @@
 {
   "$comment": "Generated by pipeline/scripts/gen-facts.mjs. Do not hand-edit: the site reads this, and a number edited here instead of at its source is the drift this file removes.",
   "generatedAt": "2026-10-13",
-  "version": "20.23.1",
+  "version": "20.24.0",
   "phaseSchema": 2,
   "phases": [
     {
@@ -35,9 +35,9 @@
     "autopilot": [0, 1, 2, 3, 4, 5],
     "analysis": [0, 1, 3, 4, 5]
   },
-  "commandCount": 64,
+  "commandCount": 65,
   "skillCount": 161,
-  "skillCountAll": 228,
+  "skillCountAll": 229,
   "agentCount": 10,
   "hosts": ["Claude Code", "Codex CLI", "Copilot CLI"],
   "toolCount": 197,