@yemi33/minions 0.1.225 → 0.1.226

package/CHANGELOG.md

@@ -1,6 +1,11 @@
 # Changelog
 
-## 0.1.225 (2026-04-03)
+## 0.1.226 (2026-04-03)
+
+### Features
+- add evaluate type to routing and playbooks (#70)
+- fix cooldown context accumulation bloat (#71)
+- wire dead test and extract progress bar helper (#64)
 
 ### Fixes
 -  silence git stderr noise during repo scan

package/dashboard/js/render-pipelines.js

@@ -64,6 +64,54 @@ function _collectRunArtifacts(run) {
   return { merged: merged, total: total };
 }
 
+/**
+ * Build a segmented progress bar from pipeline stages and a run.
+ * @param {Array} stages - pipeline stage definitions
+ * @param {Object} run - the run object containing stages status map
+ * @param {Object} [options] - optional config: { height: '8px', detailLabel: true }
+ * @returns {string} HTML string for the progress bar
+ */
+function _buildProgressBar(stages, run, options) {
+  var totalStages = stages.length;
+  var completedCount = 0;
+  var runningCount = 0;
+  var failedCount = 0;
+  stages.forEach(function(s) {
+    var st = run.stages?.[s.id]?.status;
+    if (st === 'completed') completedCount++;
+    else if (st === 'running') runningCount++;
+    else if (st === 'failed') failedCount++;
+  });
+  var pct = Math.round((completedCount / totalStages) * 100);
+
+  var segments = stages.map(function(s) {
+    var st = run.stages?.[s.id]?.status || 'pending';
+    var cls = st === 'completed' ? 'complete' : st === 'running' ? 'running' : st === 'failed' ? 'failed' : st === 'waiting-human' ? 'waiting' : 'pending';
+    return '<div class="pl-prog-seg ' + cls + '" style="width:' + (100 / totalStages) + '%" title="' + escHtml(s.id) + ': ' + st + '"></div>';
+  }).join('');
+
+  var barStyle = options && options.height ? ' style="height:' + options.height + '"' : '';
+
+  var label;
+  if (options && options.detailLabel) {
+    label = '<span style="font-weight:600;color:' + (pct === 100 ? 'var(--green)' : failedCount ? 'var(--red)' : 'var(--blue)') + '">' + pct + '% complete</span> <span style="color:var(--muted)">(' + completedCount + '/' + totalStages + ' stages)</span>';
+  } else {
+    var statusParts = [];
+    if (completedCount) statusParts.push(completedCount + ' done');
+    if (runningCount) statusParts.push(runningCount + ' running');
+    if (failedCount) statusParts.push(failedCount + ' failed');
+    var remaining = totalStages - completedCount - runningCount - failedCount;
+    if (remaining > 0) statusParts.push(remaining + ' pending');
+    label = '<span style="font-weight:600;color:' + (pct === 100 ? 'var(--green)' : failedCount ? 'var(--red)' : 'var(--blue)') + '">' + pct + '%</span>' +
+      '<span style="color:var(--muted)">' + statusParts.join(' \u00b7 ') + '</span>';
+  }
+
+  return '<div class="pl-progress-wrap">' +
+    '<div class="pl-progress-bar"' + barStyle + '>' + segments + '</div>' +
+    '<div class="pl-progress-label">' + label + '</div>' +
+  '</div>';
+}
+
 function renderPipelines(pipelines) {
   _pipelinesData = pipelines || [];
   const el = document.getElementById('pipelines-content');
@@ -95,39 +143,7 @@ function renderPipelines(pipelines) {
     var progressHtml = '';
     var displayRun = activeRun || lastRun;
     if (displayRun && (p.stages || []).length > 0) {
-      var totalStages = (p.stages || []).length;
-      var completedCount = 0;
-      var runningCount = 0;
-      var failedCount = 0;
-      (p.stages || []).forEach(function(s) {
-        var st = displayRun.stages?.[s.id]?.status;
-        if (st === 'completed') completedCount++;
-        else if (st === 'running') runningCount++;
-        else if (st === 'failed') failedCount++;
-      });
-      var pct = Math.round((completedCount / totalStages) * 100);
-
-      // Segmented progress bar — one segment per stage
-      var segments = (p.stages || []).map(function(s) {
-        var st = displayRun.stages?.[s.id]?.status || 'pending';
-        var cls = st === 'completed' ? 'complete' : st === 'running' ? 'running' : st === 'failed' ? 'failed' : st === 'waiting-human' ? 'waiting' : 'pending';
-        return '<div class="pl-prog-seg ' + cls + '" style="width:' + (100 / totalStages) + '%" title="' + escHtml(s.id) + ': ' + st + '"></div>';
-      }).join('');
-
-      var statusParts = [];
-      if (completedCount) statusParts.push(completedCount + ' done');
-      if (runningCount) statusParts.push(runningCount + ' running');
-      if (failedCount) statusParts.push(failedCount + ' failed');
-      var remaining = totalStages - completedCount - runningCount - failedCount;
-      if (remaining > 0) statusParts.push(remaining + ' pending');
-
-      progressHtml = '<div class="pl-progress-wrap">' +
-        '<div class="pl-progress-bar">' + segments + '</div>' +
-        '<div class="pl-progress-label">' +
-          '<span style="font-weight:600;color:' + (pct === 100 ? 'var(--green)' : failedCount ? 'var(--red)' : 'var(--blue)') + '">' + pct + '%</span>' +
-          '<span style="color:var(--muted)">' + statusParts.join(' \u00b7 ') + '</span>' +
-        '</div>' +
-      '</div>';
+      progressHtml = _buildProgressBar(p.stages || [], displayRun);
     }
 
     return '<div style="background:var(--surface);border:1px solid var(--border);border-radius:8px;padding:12px 16px;margin-bottom:8px;cursor:pointer" onclick="openPipelineDetail(\'' + escHtml(p.id) + '\')">' +
@@ -166,24 +182,7 @@ function openPipelineDetail(id) {
   // Stage detail with progress bar
   var detailRun = activeRun || (p.runs || []).slice(-1)[0];
   if (detailRun && (p.stages || []).length > 0) {
-    var dtotal = (p.stages || []).length;
-    var ddone = 0, drun = 0, dfail = 0;
-    (p.stages || []).forEach(function(s) {
-      var st = detailRun.stages?.[s.id]?.status;
-      if (st === 'completed') ddone++;
-      else if (st === 'running') drun++;
-      else if (st === 'failed') dfail++;
-    });
-    var dpct = Math.round((ddone / dtotal) * 100);
-    var dsegs = (p.stages || []).map(function(s) {
-      var st = detailRun.stages?.[s.id]?.status || 'pending';
-      var cls = st === 'completed' ? 'complete' : st === 'running' ? 'running' : st === 'failed' ? 'failed' : st === 'waiting-human' ? 'waiting' : 'pending';
-      return '<div class="pl-prog-seg ' + cls + '" style="width:' + (100 / dtotal) + '%" title="' + escHtml(s.id) + ': ' + st + '"></div>';
-    }).join('');
-    html += '<div class="pl-progress-wrap">' +
-      '<div class="pl-progress-bar" style="height:8px">' + dsegs + '</div>' +
-      '<div class="pl-progress-label"><span style="font-weight:600;color:' + (dpct === 100 ? 'var(--green)' : dfail ? 'var(--red)' : 'var(--blue)') + '">' + dpct + '% complete</span> <span style="color:var(--muted)">(' + ddone + '/' + dtotal + ' stages)</span></div>' +
-    '</div>';
+    html += _buildProgressBar(p.stages || [], detailRun, { height: '8px', detailLabel: true });
   }
   html += '<h4 style="font-size:12px;color:var(--blue);margin:0">Stages</h4>';
   (p.stages || []).forEach(function(s, i) {

package/engine/cooldown.js

@@ -7,10 +7,12 @@ const path = require('path');
 const shared = require('./shared');
 const queries = require('./queries');
 
+const { createHash } = require('crypto');
 const { safeJson, safeWrite, log } = shared;
 const { ENGINE_DIR } = queries;
 
 const COOLDOWN_PATH = path.join(ENGINE_DIR, 'cooldowns.json');
+const PENDING_CONTEXTS_CAP = 10;
 const dispatchCooldowns = new Map(); // key → { timestamp, failures }
 
 function loadCooldowns() {
@@ -24,6 +26,35 @@ function loadCooldowns() {
     }
   }
   log('info', `Loaded ${dispatchCooldowns.size} cooldowns from disk`);
+  // One-time purge of bloated pendingContexts on startup
+  purgeBloatedCooldowns();
+}
+
+/** Deduplicate and cap pendingContexts in all loaded cooldown entries. */
+function purgeBloatedCooldowns() {
+  let totalRemoved = 0;
+  for (const [k, v] of dispatchCooldowns) {
+    if (!Array.isArray(v.pendingContexts) || v.pendingContexts.length <= 1) continue;
+    const seen = new Set();
+    const deduped = [];
+    for (const ctx of v.pendingContexts) {
+      const hash = _contentHash(ctx);
+      if (!seen.has(hash)) {
+        seen.add(hash);
+        deduped.push(ctx);
+      }
+    }
+    const before = v.pendingContexts.length;
+    // Apply FIFO cap after dedup — keep the most recent entries
+    v.pendingContexts = deduped.length > PENDING_CONTEXTS_CAP
+      ? deduped.slice(deduped.length - PENDING_CONTEXTS_CAP)
+      : deduped;
+    totalRemoved += before - v.pendingContexts.length;
+  }
+  if (totalRemoved > 0) {
+    log('info', `Purged ${totalRemoved} duplicate/excess pendingContexts entries from cooldowns`);
+    saveCooldowns();
+  }
 }
 
 let _cooldownWriteTimer = null;
@@ -55,10 +86,26 @@ function setCooldown(key) {
   saveCooldowns();
 }
 
+function _contentHash(content) {
+  const str = typeof content === 'string' ? content : JSON.stringify(content);
+  return createHash('sha256').update(str).digest('hex');
+}
+
 function setCooldownWithContext(key, context) {
   const existing = dispatchCooldowns.get(key);
   const pendingContexts = existing?.pendingContexts || [];
-  if (context) pendingContexts.push(context);
+  if (context) {
+    // Dedup: only append if content differs from all existing entries
+    const newHash = _contentHash(context);
+    const isDuplicate = pendingContexts.some(c => _contentHash(c) === newHash);
+    if (!isDuplicate) {
+      pendingContexts.push(context);
+      // FIFO cap: drop oldest entries when exceeding cap
+      while (pendingContexts.length > PENDING_CONTEXTS_CAP) {
+        pendingContexts.shift();
+      }
+    }
+  }
   dispatchCooldowns.set(key, {
     timestamp: Date.now(),
     failures: existing?.failures || 0,
@@ -101,13 +148,16 @@ function isAlreadyDispatched(key) {
 
 module.exports = {
   COOLDOWN_PATH,
+  PENDING_CONTEXTS_CAP,
   dispatchCooldowns,
   loadCooldowns,
   saveCooldowns,
+  purgeBloatedCooldowns,
   isOnCooldown,
   setCooldown,
   setCooldownWithContext,
   getCoalescedContexts,
   setCooldownFailure,
   isAlreadyDispatched,
+  _contentHash,
 };

package/engine/playbook.js

@@ -481,7 +481,7 @@ function selectPlaybook(workType, item) {
   if (workType === 'review' && !item?._pr && !item?.pr_id) {
     return 'work-item';
   }
-  const typeSpecificPlaybooks = ['explore', 'review', 'test', 'plan-to-prd', 'plan', 'ask', 'verify', 'decompose', 'meeting-investigate', 'meeting-debate', 'meeting-conclude'];
+  const typeSpecificPlaybooks = ['explore', 'review', 'test', 'plan-to-prd', 'plan', 'ask', 'verify', 'decompose', 'evaluate', 'meeting-investigate', 'meeting-debate', 'meeting-conclude'];
   return typeSpecificPlaybooks.includes(workType) ? workType : 'work-item';
 }
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@yemi33/minions",
-  "version": "0.1.225",
+  "version": "0.1.226",
   "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/playbooks/evaluate.md

@@ -0,0 +1,149 @@
+# Playbook: Evaluate
+
+You are {{agent_name}}, the {{agent_role}} on the {{project_name}} project.
+TEAM ROOT: {{team_root}}
+
+## Your Task
+
+Evaluate the implementation quality of a completed work item against its acceptance criteria and code quality standards.
+
+## Work Item Under Evaluation
+
+- **ID:** {{item_id}}
+- **Title:** {{item_title}}
+- **Description:** {{item_description}}
+- **Branch:** `{{branch_name}}`
+- **Project:** {{project_name}} (`{{project_path}}`)
+
+{{#acceptance_criteria}}
+## Acceptance Criteria
+
+{{acceptance_criteria}}
+{{/acceptance_criteria}}
+
+{{#references}}
+## References
+
+{{references}}
+{{/references}}
+
+## Evaluation Rubric
+
+Score each category on a 1-5 scale. A category **passes** at 3 or above.
+
+### 1. Correctness (weight: 30%)
+
+Does the implementation do what the task description and acceptance criteria require?
+
+- **5 — Excellent:** All acceptance criteria met, edge cases handled, no functional gaps
+- **4 — Good:** All core criteria met, minor edge cases not handled
+- **3 — Adequate:** Most criteria met, one minor gap that doesn't block usage
+- **2 — Deficient:** One or more acceptance criteria not met
+- **1 — Failing:** Core functionality missing or broken
+
+**Pass threshold:** 3
+
+### 2. Completeness (weight: 25%)
+
+Is the implementation finished end-to-end? No TODO stubs, no half-wired features, no missing integration points.
+
+- **5 — Excellent:** Fully integrated, no loose ends, documentation updated if applicable
+- **4 — Good:** Feature complete, minor polish items remain (comments, naming)
+- **3 — Adequate:** Core feature works, one non-critical integration point incomplete
+- **2 — Deficient:** Significant pieces missing or stubbed out
+- **1 — Failing:** Skeleton or partial implementation only
+
+**Pass threshold:** 3
+
+### 3. Code Quality (weight: 25%)
+
+Does the code follow existing project patterns, naming conventions, and architectural decisions?
+
+- **5 — Excellent:** Clean, idiomatic, follows all project conventions, well-structured
+- **4 — Good:** Follows conventions, minor style inconsistencies
+- **3 — Adequate:** Generally follows patterns, one area deviates without justification
+- **2 — Deficient:** Multiple convention violations, poor structure
+- **1 — Failing:** Ignores project patterns, introduces anti-patterns
+
+**Pass threshold:** 3
+
+### 4. Test Coverage (weight: 20%)
+
+Are there tests for the new functionality? Do existing tests still pass?
+
+- **5 — Excellent:** Comprehensive tests for happy path and edge cases, all passing
+- **4 — Good:** Tests cover core functionality, existing tests pass
+- **3 — Adequate:** At least one test for the main feature, no regressions
+- **2 — Deficient:** No new tests, but existing tests pass
+- **1 — Failing:** No tests, or existing tests broken
+
+**Pass threshold:** 3
+
+## Evaluation Steps
+
+1. **Fetch and review the diff:**
+   ```bash
+   git fetch origin
+   git diff {{main_branch}}...origin/{{branch_name}}
+   ```
+
+2. **Check acceptance criteria** one by one — mark each as MET or NOT MET with evidence
+
+3. **Review code quality** — check for pattern adherence, naming, structure
+
+4. **Verify tests:**
+   ```bash
+   cd {{project_path}}
+   npm test
+   ```
+
+5. **Calculate scores** using the rubric above
+
+6. **Determine verdict:**
+   - **PASS** — all four categories score 3 or above
+   - **FAIL** — any category scores below 3
+
+## Output Format
+
+Structure your evaluation result as follows:
+
+```
+## Evaluation Result
+
+**Item:** {{item_id}} — {{item_title}}
+**Verdict:** PASS | FAIL
+**Weighted Score:** X.X / 5.0
+
+### Scores
+
+| Category | Score | Pass | Notes |
+|----------|-------|------|-------|
+| Correctness | X/5 | YES/NO | ... |
+| Completeness | X/5 | YES/NO | ... |
+| Code Quality | X/5 | YES/NO | ... |
+| Test Coverage | X/5 | YES/NO | ... |
+
+### Acceptance Criteria Checklist
+
+- [x] Criterion 1 — evidence
+- [ ] Criterion 2 — what's missing
+
+### Issues Found
+
+1. **[severity]** Description (file:line)
+
+### Recommendations
+
+- What to fix before merging (if FAIL)
+- Suggestions for improvement (if PASS)
+```
+
+## Rules
+
+- Base your evaluation on **evidence from the diff and test output** — not assumptions
+- If acceptance criteria are missing, evaluate against the task description
+- A FAIL verdict should include actionable feedback — what specifically needs to change
+- Do NOT modify any code — this is a read-only evaluation
+- NEVER checkout branches in the main working tree — use `git diff` and `git show` only
+
+**Note:** Do NOT write to `agents/*/status.json` — the engine manages your status automatically.

package/routing.md

@@ -17,6 +17,7 @@ How the engine decides who handles what. Parsed by engine.js — keep the table
 | test | dallas | ralph |
 | ask | ripley | rebecca |
 | verify | dallas | ralph |
+| evaluate | ripley | rebecca |
 | decompose | ripley | rebecca |
 | meeting | ripley | rebecca |