@yemi33/minions 0.1.353 → 0.1.355

package/CHANGELOG.md

@@ -1,8 +1,10 @@
 # Changelog
 
-## 0.1.353 (2026-04-06)
+## 0.1.355 (2026-04-06)
 
 ### Fixes
+-  pipeline plan stage includes only meeting conclusion, not full transcript
+-  pipeline plans include full meeting content, conclusion playbook requires concrete details
 -  pipeline template regex now matches hyphenated stage IDs
 -  pipeline Continue button shows immediate feedback
 -  skip meeting modal re-render when data unchanged

package/engine/pipeline.js

@@ -221,11 +221,29 @@ function executePlanStage(stage, stageState, run, config) {
   content += `**Created by:** Pipeline ${run.pipelineId}\n`;
   content += `**Date:** ${dateStamp()}\n\n---\n\n`;
 
-  // Include output from dependency stages
+  // Include output from dependency stages (for meetings, include full transcript)
   if (stage.dependsOn) {
     for (const depId of stage.dependsOn) {
       const depStage = run.stages[depId];
-      if (depStage?.output) {
+      if (!depStage) continue;
+
+      // If dependency is a meeting, include the conclusion (which should be concrete and actionable)
+      const meetingId = depStage.artifacts?.meetings?.[0];
+      if (meetingId) {
+        try {
+          const mtgPath = path.join(__dirname, '..', 'meetings', meetingId + '.json');
+          const mtg = safeJson(mtgPath);
+          if (mtg) {
+            const conclusion = typeof mtg.conclusion === 'string' ? mtg.conclusion : mtg.conclusion?.content || '';
+            if (conclusion) {
+              content += `## Meeting Conclusion: ${mtg.title || depId}\n\n${conclusion}\n\n`;
+              continue;
+            }
+          }
+        } catch { /* fall through to output-only */ }
+      }
+
+      if (depStage.output) {
         content += `## From: ${depId}\n\n${depStage.output}\n\n`;
       }
     }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@yemi33/minions",
-  "version": "0.1.353",
+  "version": "0.1.355",
   "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/meeting-conclude.md

@@ -24,14 +24,19 @@ You are {{agent_name}} ({{agent_role}}), synthesizing the team meeting results.
 
 ## Your Task
 
-Write a clear meeting conclusion covering:
+Write a clear, **self-contained** meeting conclusion. Someone reading ONLY this conclusion (not the findings or debate) must understand exactly what was decided and what to do next.
 
-1. **Areas of consensus** — what does the team agree on?
-2. **Unresolved disagreements** — where do positions still differ?
-3. **Recommended decision** — what should we do?
-4. **Action items** — specific next steps with owners (leave empty if none)
-5. **Open questions** — what still needs human input?
+1. **Top priorities** — numbered list of the specific items the team agreed to work on. For each:
+   - What the problem is (concrete: file names, function names, error messages)
+   - What the fix should be (concrete: "change X to Y", "add guard in Z")
+   - Estimated complexity (small/medium/large)
 
-Be decisive. If there's a clear best option, say so.
+2. **Deferred items** — what was discussed but explicitly deprioritized, and why
+
+3. **Unresolved disagreements** — where positions still differ (if any)
+
+4. **Open questions for human** — what needs human input before proceeding (if any)
+
+**Be concrete, not vague.** Do NOT write "fix 3 bugs" — write which 3 bugs, in which files, with what fix. The conclusion is used as input to create an implementation plan, so it must contain enough detail to act on without re-reading the entire meeting.
 
 Do NOT create plan files — plan creation is handled separately (by the pipeline plan stage or the dashboard "Create Plan" button).