claude-autopm 1.11.5 → 1.11.6

package/autopm/.claude/commands/pm/epic-decompose.md

@@ -22,29 +22,53 @@ Before proceeding, complete these validation steps.
 Do not bother the user with preflight checks progress ("I'm not going to ..."). Just do them and move on.
 
 1. **Verify epic exists:**
-   - Check if `.claude/epics/$ARGUMENTS/epic.md` exists
-   - If not found, tell user: "❌ Epic not found: $ARGUMENTS. First create it with: /pm:prd-parse $ARGUMENTS"
-   - Stop execution if epic doesn't exist
-
-2. **Check for existing tasks:**
-   - Check if any numbered task files (001.md, 002.md, etc.) already exist in `.claude/epics/$ARGUMENTS/`
+   - Check if `.claude/epics/$ARGUMENTS` directory exists
+   - Check for either:
+     a) Single epic: `.claude/epics/$ARGUMENTS/epic.md` exists
+     b) Multiple epics: Subdirectories like `.claude/epics/$ARGUMENTS/01-infrastructure/epic.md`
+   - If neither found, tell user: "❌ Epic not found: $ARGUMENTS. First create it with: /pm:prd-parse $ARGUMENTS or /pm:epic-split $ARGUMENTS"
+   - Stop execution if no epics found
+
+2. **Detect epic structure:**
+   - If `.claude/epics/$ARGUMENTS/epic.md` exists → Single epic mode
+   - If subdirectories with epic.md files exist → Multi-epic mode
+   - Store the mode for later processing
+
+3. **Check for existing tasks:**
+   - For single epic: Check `.claude/epics/$ARGUMENTS/` for numbered task files
+   - For multi-epic: Check each subdirectory for numbered task files
    - If tasks exist, list them and ask: "⚠️ Found {count} existing tasks. Delete and recreate all tasks? (yes/no)"
    - Only proceed with explicit 'yes' confirmation
    - If user says no, suggest: "View existing tasks with: /pm:epic-show $ARGUMENTS"
 
-3. **Validate epic frontmatter:**
-   - Verify epic has valid frontmatter with: name, status, created, prd
-   - If invalid, tell user: "❌ Invalid epic frontmatter. Please check: .claude/epics/$ARGUMENTS/epic.md"
+4. **Validate epic frontmatter:**
+   - For single epic: Verify `.claude/epics/$ARGUMENTS/epic.md` has valid frontmatter
+   - For multi-epic: Verify each subdirectory's epic.md has valid frontmatter
+   - If invalid, tell user which epic file has invalid frontmatter
 
-4. **Check epic status:**
-   - If epic status is already "completed", warn user: "⚠️ Epic is marked as completed. Are you sure you want to decompose it again?"
+5. **Check epic status:**
+   - For each epic, check if status is already "completed"
+   - If any epic is completed, warn user: "⚠️ Epic(s) marked as completed. Are you sure you want to decompose again?"
 
 ## Instructions
 
-You are decomposing an epic into specific, actionable tasks for: **$ARGUMENTS**
+You are decomposing epic(s) into specific, actionable tasks for: **$ARGUMENTS**
+
+### 1. Determine Processing Mode
+
+**Single Epic Mode:**
+- Process `.claude/epics/$ARGUMENTS/epic.md`
+- Create tasks in `.claude/epics/$ARGUMENTS/`
 
-### 1. Read the Epic
-- Load the epic from `.claude/epics/$ARGUMENTS/epic.md`
+**Multi-Epic Mode (from epic-split):**
+- Find all subdirectories in `.claude/epics/$ARGUMENTS/`
+- Process each subdirectory's epic.md file separately
+- Create tasks in each respective subdirectory
+- Show progress for each epic being processed
+
+### 2. Read the Epic(s)
+- For single epic: Load from `.claude/epics/$ARGUMENTS/epic.md`
+- For multi-epic: Load each `.claude/epics/$ARGUMENTS/*/epic.md`
 - Understand the technical approach and requirements
 - Review the task breakdown preview
 
@@ -127,9 +151,17 @@ Clear, concise description of what needs to be done
 ```
 
 ### 3. Task Naming Convention
+
+**For Single Epic:**
 Save tasks as: `.claude/epics/$ARGUMENTS/{task_number}.md`
+
+**For Multi-Epic:**
+Save tasks as: `.claude/epics/$ARGUMENTS/{epic_folder}/{task_number}.md`
+Example: `.claude/epics/ecommerce/01-infrastructure/001.md`
+
 - Use sequential numbering: 001.md, 002.md, etc.
 - Keep task titles short but descriptive
+- Each epic gets its own task number sequence
 
 ### 4. Frontmatter Guidelines
 - **name**: Use a descriptive task title (without "Task:" prefix)
@@ -178,6 +210,23 @@ Spawning 3 agents for parallel task creation:
 - Agent 3: Creating tasks 007-009 (UI layer)
 ```
 
+**Multi-Epic Processing Example:**
+```markdown
+Processing multiple epics from split:
+
+📂 01-infrastructure/epic.md
+   Creating 8 tasks...
+   ✅ Done
+
+📂 02-auth-backend/epic.md
+   Creating 12 tasks...
+   ✅ Done
+
+📂 03-frontend/epic.md
+   Creating 10 tasks...
+   ✅ Done
+```
+
 ### 8. Task Dependency Validation
 
 When creating tasks with dependencies:
@@ -212,6 +261,7 @@ Before finalizing tasks, verify:
 
 ### 10. Post-Decomposition
 
+**For Single Epic:**
 After successfully creating tasks:
 1. Confirm: "✅ Created {count} tasks for epic: $ARGUMENTS"
 2. Show summary:
@@ -220,6 +270,24 @@ After successfully creating tasks:
    - Total estimated effort
 3. Suggest next step: "Ready to sync to GitHub? Run: /pm:epic-sync $ARGUMENTS"
 
+**For Multi-Epic:**
+After processing all epics:
+1. Show per-epic summary:
+   ```
+   ✅ Epic Decomposition Complete
+
+   📂 01-infrastructure: 8 tasks created
+   📂 02-auth-backend: 12 tasks created
+   📂 03-frontend: 10 tasks created
+
+   Total: 30 tasks across 3 epics
+   ```
+2. Show combined statistics:
+   - Total tasks across all epics
+   - Total estimated effort
+   - Breakdown by epic
+3. Suggest next step: "Ready to sync all epics? Run: /pm:epic-sync $ARGUMENTS"
+
 ## Error Recovery
 
 If any step fails:

package/autopm/.claude/commands/pm/epic-sync.md

@@ -14,11 +14,18 @@ Push epic and tasks to GitHub as issues using modular scripts.
 ## Quick Check
 
 ```bash
-# Verify epic exists
-test -f .claude/epics/$ARGUMENTS/epic.md || echo "❌ Epic not found. Run: /pm:prd-parse $ARGUMENTS"
-
-# Count task files
-ls .claude/epics/$ARGUMENTS/*.md 2>/dev/null | grep -v epic.md | wc -l
+# Check for single epic or multi-epic structure
+if [ -f ".claude/epics/$ARGUMENTS/epic.md" ]; then
+    echo "Single epic mode detected"
+    # Count task files
+    ls .claude/epics/$ARGUMENTS/*.md 2>/dev/null | grep -v epic.md | wc -l
+elif [ -d ".claude/epics/$ARGUMENTS" ] && ls .claude/epics/$ARGUMENTS/*/epic.md 2>/dev/null | head -1; then
+    echo "Multi-epic mode detected (from epic-split)"
+    # Count task files in all subdirectories
+    find .claude/epics/$ARGUMENTS -name "*.md" ! -name "epic.md" ! -name "meta.yaml" | wc -l
+else
+    echo "❌ Epic not found. Run: /pm:prd-parse $ARGUMENTS or /pm:epic-split $ARGUMENTS"
+fi
 ```
 
 If no tasks found: "❌ No tasks to sync. Run: /pm:epic-decompose $ARGUMENTS"
@@ -27,6 +34,18 @@ If no tasks found: "❌ No tasks to sync. Run: /pm:epic-decompose $ARGUMENTS"
 
 The epic sync process is now modularized into 4 specialized scripts that handle different aspects of the synchronization. Each script is designed for reliability, testability, and maintainability.
 
+### Processing Mode Detection
+
+**Single Epic Mode:**
+- Process `.claude/epics/$ARGUMENTS/epic.md` and its tasks
+- Create one epic issue with all tasks linked
+
+**Multi-Epic Mode (from epic-split):**
+- Process each subdirectory separately
+- Create separate epic issues for each subdirectory
+- Maintain epic dependencies as specified in meta.yaml
+- Show progress for each epic being synced
+
 ### 1. Repository Protection Check
 
 This is handled automatically by our modular scripts, but you can run the check manually:
@@ -137,9 +156,11 @@ git push -u origin epic/$ARGUMENTS
 echo "✅ Created branch: epic/$ARGUMENTS"
 ```
 
-## Complete Workflow Example
+## Complete Workflow Examples
 
-Here's the complete modular epic sync workflow:
+### Single Epic Workflow
+
+Here's the complete modular epic sync workflow for a single epic:
 
 ```bash
 #!/bin/bash
@@ -215,6 +236,100 @@ echo "  - Or work on single issue: /pm:issue-start <issue_number>"
 echo ""
 ```
 
+### Multi-Epic Workflow (from epic-split)
+
+Here's the workflow for syncing multiple epics created by epic-split:
+
+```bash
+#!/bin/bash
+# Sync multiple epics from split structure
+
+FEATURE_NAME="$ARGUMENTS"
+EPICS_DIR=".claude/epics/$FEATURE_NAME"
+
+echo "🚀 Starting multi-epic sync for: $FEATURE_NAME"
+
+# Find all epic subdirectories
+epic_dirs=$(find "$EPICS_DIR" -maxdepth 1 -type d -name "[0-9]*-*" | sort)
+epic_count=$(echo "$epic_dirs" | wc -l)
+
+echo "📦 Found $epic_count epics to sync"
+
+# Track all created epic issues
+all_epic_numbers=""
+total_tasks=0
+
+# Process each epic
+for epic_dir in $epic_dirs; do
+    epic_name=$(basename "$epic_dir")
+    echo ""
+    echo "━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"
+    echo "📂 Processing: $epic_name"
+    echo "━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"
+
+    # Step 1: Create epic issue
+    echo "📝 Creating epic issue..."
+    epic_number=$(bash autopm/.claude/scripts/pm/epic-sync/create-epic-issue.sh "$FEATURE_NAME/$epic_name")
+
+    if [[ -z "$epic_number" ]]; then
+        echo "⚠️  Failed to create epic issue for $epic_name"
+        continue
+    fi
+
+    all_epic_numbers="$all_epic_numbers $epic_number"
+    echo "✅ Epic issue created: #$epic_number"
+
+    # Step 2: Create task issues for this epic
+    echo "📋 Creating task issues..."
+    task_mapping_file=$(bash autopm/.claude/scripts/pm/epic-sync/create-task-issues.sh "$FEATURE_NAME/$epic_name" "$epic_number")
+
+    if [[ -f "$task_mapping_file" ]]; then
+        task_count=$(wc -l < "$task_mapping_file")
+        total_tasks=$((total_tasks + task_count))
+        echo "✅ Created $task_count task issues"
+
+        # Step 3: Update references
+        echo "🔗 Updating task references..."
+        bash autopm/.claude/scripts/pm/epic-sync/update-references.sh "$FEATURE_NAME/$epic_name" "$task_mapping_file"
+
+        # Step 4: Update epic file
+        bash autopm/.claude/scripts/pm/epic-sync/update-epic-file.sh "$FEATURE_NAME/$epic_name" "$epic_number"
+    fi
+done
+
+# Create meta-branch for all epics
+echo ""
+echo "🌿 Creating meta-branch for feature..."
+git checkout main
+git pull origin main
+git checkout -b feature/$FEATURE_NAME
+git push -u origin feature/$FEATURE_NAME
+
+# Final summary
+repo=$(gh repo view --json nameWithOwner -q .nameWithOwner)
+
+echo ""
+echo "🎉 Multi-epic sync completed!"
+echo ""
+echo "📊 Summary:"
+echo "  Feature: $FEATURE_NAME"
+echo "  Epics created: $epic_count"
+echo "  Epic issues:$all_epic_numbers"
+echo "  Total tasks: $total_tasks"
+echo "  Branch: feature/$FEATURE_NAME"
+echo ""
+echo "🔗 Links:"
+for epic_num in $all_epic_numbers; do
+    echo "  Epic #$epic_num: https://github.com/$repo/issues/$epic_num"
+done
+echo "  Branch: https://github.com/$repo/tree/feature/$FEATURE_NAME"
+echo ""
+echo "📋 Next steps:"
+echo "  - Work on P0 epics first (infrastructure, auth backend)"
+echo "  - Start specific epic: /pm:issue-start <issue_number>"
+echo ""
+```
+
 ## Benefits of Modular Approach
 
 ### ✅ **Reliability**

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "claude-autopm",
-  "version": "1.11.5",
+  "version": "1.11.6",
   "description": "Autonomous Project Management Framework for Claude Code - Advanced AI-powered development automation",
   "main": "bin/autopm.js",
   "bin": {