@yemi33/minions 0.1.1036 → 0.1.1038

package/CHANGELOG.md

@@ -1,5 +1,90 @@
 # Changelog
 
+## 0.1.1038 (2026-04-16)
+
+### Features
+-  Doc-chat performance — debounced persistence, session TTL, smart disk reads (#1164)
+- split getStatus() into fast/slow state tiers with separate TTLs (#1170)
+-  Cache _countWorktrees() with 30s TTL (#1166)
+- Pre-cache gzipped status buffer alongside JSON cache (#1171)
+- parallelize ADO and GitHub PR polling with Promise.allSettled (#1172)
+- route implement items to dedicated implement playbook (#1115)
+
+### Fixes
+- allow test tasks to create PRs when files are modified
+-  sidecar oversized dispatch prompts to prevent dashboard OOM (closes #1167) (#1183)
+-  use locked write for failed-with-PR reconciliation in cleanup.js (#1169)
+-  prevent test from corrupting live meeting state
+- scheduler enabled field falsy check treats undefined as disabled (#1160)
+-  update branch-dedup tests to use canonical PR IDs after merge with master (#1146)
+-  auto-review not firing for manually-linked PRs with autoObserve=true
+-  mark PR abandoned on 404 instead of silently retrying each tick
+- replace undefined PROJECTS with config.projects in checkWatches (#1108)
+- write permission for publish workflow
+-  run tests inline and post check runs for publish PRs
+- add maxBuffer to all GitHub CLI spawns + paginate PR list (#1130)
+-  add required CI checks for PRs + update publish for auto-merge
+-  revert to PR merge now that stale status check is removed
+- guard live review check against undefined vote/state values (#1132)
+-  push version bump directly to master instead of via PR
+-  add push-triggered CI for chore/publish branches
+-  publish workflow chore PRs failing to merge
+-  harden KB ordering
+-  harden audited state transitions
+
+### Other
+- Fix audit cleanup and test isolation
+- docs(sched-weekly-docs-cleanup-1776355200664): weekly documentation cleanup (#1181)
+- [E2E] Dashboard & engine perf: gzip cache, tiered status, lock backoff, polling parallelization (#1174)
+- Fix doc chat session isolation
+- test(pipeline): add unit tests for CRUD, stage execution, run lifecycle (#1162)
+- chore: test publish after removing stale status check
+- chore: trigger publish test
+- chore: test publish workflow fix
+- chore: trigger publish workflow test
+
+## 0.1.1037 (2026-04-16)
+
+### Features
+-  Doc-chat performance — debounced persistence, session TTL, smart disk reads (#1164)
+- split getStatus() into fast/slow state tiers with separate TTLs (#1170)
+-  Cache _countWorktrees() with 30s TTL (#1166)
+- Pre-cache gzipped status buffer alongside JSON cache (#1171)
+- parallelize ADO and GitHub PR polling with Promise.allSettled (#1172)
+- route implement items to dedicated implement playbook (#1115)
+
+### Fixes
+- allow test tasks to create PRs when files are modified
+-  sidecar oversized dispatch prompts to prevent dashboard OOM (closes #1167) (#1183)
+-  use locked write for failed-with-PR reconciliation in cleanup.js (#1169)
+-  prevent test from corrupting live meeting state
+- scheduler enabled field falsy check treats undefined as disabled (#1160)
+-  update branch-dedup tests to use canonical PR IDs after merge with master (#1146)
+-  auto-review not firing for manually-linked PRs with autoObserve=true
+-  mark PR abandoned on 404 instead of silently retrying each tick
+- replace undefined PROJECTS with config.projects in checkWatches (#1108)
+- write permission for publish workflow
+-  run tests inline and post check runs for publish PRs
+- add maxBuffer to all GitHub CLI spawns + paginate PR list (#1130)
+-  add required CI checks for PRs + update publish for auto-merge
+-  revert to PR merge now that stale status check is removed
+- guard live review check against undefined vote/state values (#1132)
+-  push version bump directly to master instead of via PR
+-  add push-triggered CI for chore/publish branches
+-  publish workflow chore PRs failing to merge
+-  harden KB ordering
+-  harden audited state transitions
+
+### Other
+- docs(sched-weekly-docs-cleanup-1776355200664): weekly documentation cleanup (#1181)
+- [E2E] Dashboard & engine perf: gzip cache, tiered status, lock backoff, polling parallelization (#1174)
+- Fix doc chat session isolation
+- test(pipeline): add unit tests for CRUD, stage execution, run lifecycle (#1162)
+- chore: test publish after removing stale status check
+- chore: trigger publish test
+- chore: test publish workflow fix
+- chore: trigger publish workflow test
+
 ## 0.1.1036 (2026-04-16)
 
 ### Features

package/agents/lambert/charter.md

@@ -14,7 +14,7 @@
 - Creating the structured PRD in JSON format
 - Gap analysis: what was built vs. what the codebase instructions describe vs. what's missing
 - Feature inventory: cataloguing existing agent capabilities, then identifying what's absent
-- Producing `docs/prd-gaps.json` with structured feature records
+- Producing PRD JSON files in `{{team_root}}/prd/` with structured feature records (one file per plan; filename matches the source plan slug)
 - Flagging unknowns and ambiguities in the requirements
 
 ## PRD JSON Output Format
@@ -43,7 +43,7 @@
 - Read Dallas's build summary from `{{team_root}}/notes/inbox/dallas-build-*.md`
 - Cross-reference against all `docs/`, agent `CLAUDE.md` files, and prototype instructions
 - Be exhaustive on missing features — better to over-document than under-document
-- Output the JSON to `docs/prd-gaps.json`
+- Output the JSON to `{{team_root}}/prd/<plan-slug>.json` (the engine's `materializePlansAsWorkItems()` reads from that directory)
 
 ## Boundaries
 

package/docs/auto-discovery.md

@@ -37,13 +37,13 @@ Skips PRs where `status !== "active"`.
 
 ### Source 2: PRD Gap Analysis (via `materializePlansAsWorkItems`)
 
-PRD items flow through `materializePlansAsWorkItems()`, which scans `~/.minions/prd/*.json` for PRD files with `missing`/`planned` items and creates work items in the target project's queue.
+PRD items flow through `materializePlansAsWorkItems()`, which scans `~/.minions/prd/*.json` for PRD files with `missing` / `updated` / `planned` items and creates work items in the target project's queue.
 
-**Reads:** `<project>/docs/prd-gaps.json` (or custom path from `workSources.prd.path`)
+**Reads:** `~/.minions/prd/*.json` — the central PRD directory. (Legacy `<project>/docs/prd-gaps.json` paths are no longer scanned.)
 
 | Item State | Action | Dispatch Type |
 |------------|--------|---------------|
-| `status: "missing"` or `"planned"` | Queue implementation | `implement` |
+| `status: "missing"` or `"updated"` | Queue implementation (re-opens any existing done WI) | `implement` |
 | `estimated_complexity: "large"` | Routes to `implement:large` (prefers Rebecca) | `implement:large` |
 
 ### Source 3: Per-Project Work Items (`discoverFromWorkItems`)
@@ -165,14 +165,19 @@ Item found
 `resolveAgent()` parses `routing.md` to pick the right agent:
 
 ```
-routing.md table:
+routing.md table (see the file for the authoritative list):
   implement       → dallas  (fallback: ralph)
   implement:large → rebecca (fallback: dallas)
   review          → ripley  (fallback: lambert)
-  fix             → _author_ (fallback: dallas)   ← routes to PR author
-  analyze         → lambert (fallback: rebecca)
+  fix             → _author_ (fallback: _any_)    ← routes to PR author, any idle as fallback
+  plan            → ripley  (fallback: rebecca)
+  plan-to-prd     → lambert (fallback: rebecca)
   explore         → ripley  (fallback: rebecca)
   test            → dallas  (fallback: ralph)
+  ask             → ripley  (fallback: rebecca)
+  verify          → dallas  (fallback: ralph)
+  decompose       → ripley  (fallback: rebecca)
+  meeting         → ripley  (fallback: lambert)
 ```
 
 Resolution order:
@@ -216,7 +221,7 @@ git worktree add <rootDir>/../worktrees/<branch> -b <branch> <mainBranch>
 ```
 - Branch names are sanitized (alphanumeric, dots, hyphens, slashes only, max 200 chars)
 - If worktree fails for implement/fix → item marked as error, moved to completed
-- If worktree fails for review/analyze → falls back to rootDir (read-only tasks)
+- If worktree fails for review/explore/ask → falls back to rootDir (read-only tasks)
 
 ### 3. Render Playbook
 ```
@@ -291,8 +296,8 @@ The dashboard exposes a unified input box at `http://localhost:7331` that parses
 | `@agent` | Assigns to a specific agent (sets `item.agent`) |
 | `@everyone` | Fan-out to all agents (sets `scope: 'fan-out'`) |
 | `!high` / `!low` | Sets priority (default: medium) |
-| `/decide` | Creates a decision (appended to `notes.md`) |
-| `/prd` | Creates a PRD item (appended to `prd-gaps.json`) |
+| `/note` | Writes a note to `notes/inbox/` for consolidation into `notes.md` |
+| `/plan` | Creates a plan draft (appended to `plans/`) |
 | `#project` | Targets a specific project queue |
 
 Work type is auto-detected from keywords (fix, explore, test, review → implement as fallback). The `@agent` assignment flows through to the engine: `item.agent || resolveAgent(workType, config)`.
@@ -305,20 +310,20 @@ Work type is auto-detected from keywords (fix, explore, test, review → impleme
                               │
                    ┌──────────┼──────────┐
                    ▼          ▼          ▼
-             work-items   decisions   prd-gaps
-             .json        .md         .json
+             work-items   notes/       plans/
+             .json        inbox/*.md   *.md
 
 Per-project sources:              Central engine:              Agents:
 
 work-items.json ──┐
-prd-gaps.json ────┤
 pull-requests.json┤  discoverWork()   dispatch.json
 docs/**/*.md (specs)┤  (each tick)      ┌──────────┐
                   │       │           │ pending   │
 ~/.minions/         │       │           │ active    │
   work-items.json ┤       │           │ completed │
-  plans/*.md ─────┘       ▼           └─────┬────┘
-                     addToDispatch()─────────┘
+  prd/*.json ─────┤       │           └─────┬────┘
+  plans/*.md ─────┘       ▼                 │
+                     addToDispatch()────────┘
                                             │
 ADO REST API ─── pollPrBuildStatus() ──► pull-requests.json
 (every 6min)      (buildStatus field)       │
@@ -390,20 +395,12 @@ All discovery behavior is controlled via `config.json`:
       "name": "MyProject",
       "localPath": "C:/Users/you/MyProject",
       "workSources": {
-        "prd": {
-          "enabled": true,
-          "path": "docs/prd-gaps.json",
-          "itemFilter": { "status": ["missing", "planned"] },
-          "cooldownMinutes": 30
-        },
         "pullRequests": {
           "enabled": true,
-          "path": "projects/<name>/pull-requests.json",
           "cooldownMinutes": 30
         },
         "workItems": {
           "enabled": true,
-          "path": "projects/<name>/work-items.json",
           "cooldownMinutes": 0
         }
       }

package/docs/command-center.md

@@ -11,9 +11,9 @@ Click the **CC** button in the top-right header of the dashboard. A slide-out ch
 CC maintains a true multi-turn session using Claude CLI's `--resume` flag. Unlike a typical stateless API call, each message resumes the same conversation — Claude retains full history including tool calls, intermediate reasoning, and file contents from prior turns.
 
 **Session lifecycle:**
-- **Created** on first message (or after expiry/new-session)
+- **Created** on first message (or after the system prompt changes, or when you click **New Session**)
 - **Resumed** on subsequent messages via `--resume <sessionId>`
-- **Expires** after 2 hours of inactivity
+- **Invalidated** only when the CC system prompt changes — detected by hashing `CC_STATIC_SYSTEM_PROMPT` into `_ccPromptHash` and comparing on each call. There is no time-based expiry and no turn cap (`CC_SESSION_MAX_TURNS = Infinity`).
 - **Persisted** to `engine/cc-session.json` — survives dashboard restarts
 - **Frontend messages** saved to `localStorage` — survive page refresh
 
@@ -108,7 +108,7 @@ User message (CC panel, doc modal, or steer)
     ▼
 POST /api/command-center  (or /api/doc-chat, /api/steer-document)
     │
-    ├── Validate session (expiry check)
+    ├── Validate session (prompt-hash check — invalidate if `_ccPromptHash` drifted)
     ├── Build dynamic state preamble (buildCCStatePreamble)
     ├── callLLM() with sessionId (resume) or without (new)
     │     model: sonnet

package/engine/cleanup.js

@@ -448,17 +448,19 @@ function runCleanup(config, verbose = false) {
     try {
       const wiPath = projectWorkItemsPath(project);
       let reconciled = 0;
-      mutateWorkItems(wiPath, items => {
-        for (const item of items) {
-          if (item.status === shared.WI_STATUS.FAILED && item._pr) {
-            item.status = shared.WI_STATUS.DONE;
-            if (item.failReason) delete item.failReason;
-            if (item.failedAt) delete item.failedAt;
-            if (!item.completedAt) item.completedAt = shared.ts();
-            reconciled++;
+        mutateWorkItems(wiPath, items => {
+          for (const item of items) {
+            if (item.status === shared.WI_STATUS.FAILED && item._pr) {
+              item.status = shared.WI_STATUS.DONE;
+              if (item.failReason) delete item.failReason;
+              if (item.failedAt) delete item.failedAt;
+              delete item._retryCount;
+              delete item._pendingReason;
+              if (!item.completedAt) item.completedAt = shared.ts();
+              reconciled++;
+            }
           }
-        }
-      });
+        });
       if (reconciled > 0) {
         log('info', `Reconciled ${reconciled} failed-with-PR item(s) → done in ${project.name}`);
       }
@@ -472,15 +474,17 @@ function runCleanup(config, verbose = false) {
     try {
       const wiPath = projectWorkItemsPath(project);
       let migrated = 0;
-      mutateWorkItems(wiPath, items => {
-        for (const item of items) {
-          if (LEGACY_DONE_ALIASES.has(item.status)) {
-            item.status = shared.WI_STATUS.DONE;
-            delete item._pendingReason;
-            migrated++;
+        mutateWorkItems(wiPath, items => {
+          for (const item of items) {
+            if (LEGACY_DONE_ALIASES.has(item.status)) {
+              item.status = shared.WI_STATUS.DONE;
+              delete item._retryCount;
+              delete item._pendingReason;
+              if (!item.completedAt) item.completedAt = shared.ts();
+              migrated++;
+            }
           }
-        }
-      });
+        });
       if (migrated > 0) {
         log('info', `Migrated ${migrated} legacy status(es) → done in ${project.name} work items`);
       }
@@ -494,7 +498,9 @@ function runCleanup(config, verbose = false) {
       for (const item of items) {
         if (LEGACY_DONE_ALIASES.has(item.status)) {
           item.status = shared.WI_STATUS.DONE;
+          delete item._retryCount;
           delete item._pendingReason;
+          if (!item.completedAt) item.completedAt = shared.ts();
           migrated++;
         }
       }
@@ -503,6 +509,36 @@ function runCleanup(config, verbose = false) {
       log('info', `Migrated ${migrated} legacy status(es) → done in central work items`);
     }
   } catch (e) { log('warn', 'migrate central legacy statuses: ' + e.message); }
+
+  // 6c. Strip stale retry metadata from completed work items
+  cleaned.doneRetryCounts = 0;
+  for (const project of projects) {
+    try {
+      const wiPath = projectWorkItemsPath(project);
+      mutateWorkItems(wiPath, items => {
+        for (const item of items) {
+          if (item.status === shared.WI_STATUS.DONE && item._retryCount !== undefined) {
+            delete item._retryCount;
+            cleaned.doneRetryCounts++;
+          }
+        }
+      });
+    } catch (e) { log('warn', 'cleanup done retry metadata: ' + e.message); }
+  }
+  try {
+    const centralPath = path.join(MINIONS_DIR, 'work-items.json');
+    mutateWorkItems(centralPath, items => {
+      for (const item of items) {
+        if (item.status === shared.WI_STATUS.DONE && item._retryCount !== undefined) {
+          delete item._retryCount;
+          cleaned.doneRetryCounts++;
+        }
+      }
+    });
+  } catch (e) { log('warn', 'cleanup central done retry metadata: ' + e.message); }
+  if (cleaned.doneRetryCounts > 0) {
+    log('info', `Cleanup: cleared ${cleaned.doneRetryCounts} stale retry count(s) from done work items`);
+  }
   // PRD items (missing_features[].status)
   try {
     let prdDirEntries;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@yemi33/minions",
-  "version": "0.1.1036",
+  "version": "0.1.1038",
   "description": "Multi-agent AI dev team that runs from ~/.minions/ — five autonomous agents share a single engine, dashboard, and knowledge base",
   "bin": {
     "minions": "bin/minions.js"

package/docs/verify-minions-2026-04-01-5.md

@@ -1,32 +0,0 @@
-# Verification: Fix silent state-loss bugs in dispatch persistence, lifecycle races, and checkpoint handling
-
-**Plan:** minions-2026-04-01-5.json
-**Date:** 2026-04-11
-**Verified by:** Ripley
-
-## Results Summary
-
-| Item | PR | Verdict |
-|------|-----|---------|
-| P-b3c9e4f6: needs-human-review skip filter | PR-55 (merged) | PASS |
-| P-f7a2d8e1: dispatch persistence gate | PR-57 (merged) | PASS |
-| P-d5e1a7b3: TOCTOU races in lifecycle | PR-56 (merged) | PASS |
-| P-a8f4c2d9: fan-out checkpoint handling | PR-58 (merged) | REGRESSION |
-| P-e6b0d3a5: checkpoint cleanup + _pendingReason | PR-60 (merged) | REGRESSION |
-| P-c1d7f9e8: resolveWiPath export | PR-59 (closed) | NOT MERGED |
-
-**Tests:** 1463 passed, 0 failed, 2 skipped
-
-## Regressions
-
-Two items (P-a8f4c2d9, P-e6b0d3a5) were correctly merged but their changes were
-subsequently lost during engine.js refactors that rewrote `discoverCentralWorkItems`
-to use a mutations Map pattern. The original direct-item-mutation code was not carried
-forward.
-
-Affected features:
-1. Fan-out checkpoint.json reading (engine.js:2484-2487 missing `worktreePath`)
-2. Checkpoint file deletion after consumption (no `unlinkSync` calls remain)
-3. `_pendingReason` clearing on fan-out/single-agent dispatch
-
-See full report in `prd/guides/verify-minions-2026-04-01-5.md` (runtime, gitignored).