ldy26098
/
BMAD-METHOD
mirror of https://github.com/bmad-code-org/BMAD-METHOD.git
1
0
Fork 0
Code Packages Projects Releases Wiki Activity
BMAD-METHOD/expansion-packs/story-implementation/tasks/commit-and-prepare-pr.md

7.5 KiB
Raw Blame History

Commit and Prepare PR

Task Overview

Agent: dev
Action Type: git-commit-and-pr-preparation
Duration: 5-10 minutes
LLM-Optimized: Structured commit and context generation

Purpose

Commit all story implementation changes with comprehensive context and prepare detailed PR context for comprehensive review and delivery.

Inputs

  • Story implementation file with complete learning review results
  • All implementation code changes
  • Test results and quality gate confirmations
  • Review feedback and learning items

Outputs

  • Git commit with story implementation changes
  • Comprehensive PR context prepared
  • Story file updated with commit information
  • Ready state for PR creation

Execution Steps

Step 1: Pre-Commit Validation (2 minutes)

PRE_COMMIT_CHECKLIST:
- [ ] All quality gates passing
- [ ] Story file updated with learning review results  
- [ ] Implementation code complete and tested
- [ ] No uncommitted changes remaining
- [ ] Branch synchronized with latest main/develop

Step 2: Generate Commit Message (2 minutes)

COMMIT_MESSAGE_FORMAT:
[Epic-Story] Brief implementation summary

Story: Epic {epic_number}, Story {story_number}
Type: [feature/enhancement/fix/refactor]

Implementation Summary:
- [Key change 1 - max 50 tokens]
- [Key change 2 - max 50 tokens]  
- [Key change 3 - max 50 tokens]

Quality Gates: [PASS/FAIL counts]
Review Rounds: [1/2/3+]
Learning Items: [X items captured]

Business Value: [Brief description - max 100 tokens]
Technical Impact: [Brief description - max 100 tokens]

Generated with Claude Code
Co-Authored-By: Claude <noreply@anthropic.com>

Step 3: Commit Implementation (1 minute)

# Add all story-related changes
git add .

# Create commit with comprehensive message
git commit -m "$(cat <<'EOF'
[Epic{epic_number}-Story{story_number}] {brief_summary}

Story: Epic {epic_number}, Story {story_number}
Type: {change_type}

Implementation Summary:
- {key_change_1}
- {key_change_2}
- {key_change_3}

Quality Gates: {pass_count} PASS, {fail_count} FAIL
Review Rounds: {review_rounds}
Learning Items: {learning_count} items captured

Business Value: {business_value_summary}
Technical Impact: {technical_impact_summary}

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
EOF
)"

Step 4: Prepare PR Context (3-5 minutes)

Generate comprehensive PR context document:

# PR Context: Epic {epic_number}.{story_number}

## Business Summary
**Epic:** {epic_title}
**Epic Progress:** {epic_completion_percentage}% complete ({completed_stories}/{total_stories} stories)
**Story:** {story_title}
**Type:** {change_type}
**Complexity:** {SIMPLE/MODERATE/COMPLEX}
**Epic Status:** {IN_PROGRESS/COMPLETE}
**Epic Retrospective:** {MANDATORY_AUTO_TRIGGERED/PENDING/NOT_REQUIRED}

### Epic Completion Status
**If Epic Complete (100%):**
- 🎉 **EPIC COMPLETION ACHIEVED!** Epic {epic_number} is now 100% complete
- 📊 **Epic Retrospective:** MANDATORY and automatically triggered
- 🎆 **Epic Celebration:** Multi-agent retrospective scheduled for strategic insights
- 🎣 **Next Epic Preparation:** Action items will be generated during retrospective

**If Epic In Progress (<100%):**
- 🚧 **Epic Progress:** {epic_completion_percentage}% complete, {remaining_stories} stories remaining
- 📅 **Next Story:** Story {next_story_number} ready for development
- 🔄 **Epic Timeline:** On track for completion by {projected_completion_date}

### Business Value
- {business_impact_1}
- {business_impact_2}
- {business_impact_3}

## Technical Changes
### Implementation Summary
- {technical_change_1} | Impact: {HIGH/MEDIUM/LOW}
- {technical_change_2} | Impact: {HIGH/MEDIUM/LOW}
- {technical_change_3} | Impact: {HIGH/MEDIUM/LOW}

### Quality Metrics
- **Tests:** {test_count} added, {existing_test_count} updated
- **Code Coverage:** {coverage_percentage}%
- **Quality Gates:** {pass_count} PASS, {fail_count} FAIL
- **Review Rounds:** {review_rounds}

### Architecture Impact
- {architecture_impact_1}
- {architecture_impact_2}

## Learning Extraction
### Immediate Actions (Current Sprint)
- {immediate_action_1} - {owner} - Due: {date}
- {immediate_action_2} - {owner} - Due: {date}

### Next Sprint Integration
- {next_sprint_action_1} - {owner}
- {next_sprint_action_2} - {owner}

### Future Epic Candidates
- {epic_candidate_1} - Priority: {HIGH/MEDIUM/LOW}
- {epic_candidate_2} - Priority: {HIGH/MEDIUM/LOW}

### Epic Retrospective Context (if Epic Complete)
**Epic Retrospective Data Prepared:**
- All {total_stories} story files consolidated
- {total_learning_items} learning items across epic
- Epic metrics: {avg_quality_score}/10 quality, {epic_duration} days duration
- Multi-agent retrospective scheduled with: SM (facilitator), Architect, PO, Dev, UX-Expert
- Strategic insights and next epic preparation action items to be generated

**Epic Retrospective Status:** {MANDATORY_TRIGGERED/NOT_APPLICABLE}

## Validation Evidence
### Pre-Review Validation
- {validation_item_1}: PASS
- {validation_item_2}: PASS
- {validation_item_3}: PASS

### Review Results
- **Architecture Review:** {PASS/ADDRESSED}
- **Business Review:** {PASS/ADDRESSED}
- **QA Review:** {PASS/ADDRESSED}
- **UX Review:** {PASS/ADDRESSED}

### Final Validation
- **Quality Gates:** ALL PASS
- **Story DoD:** COMPLETE
- **Learning Extraction:** COMPLETE

## Files Changed
- {file_1} - {change_type} - {line_count} lines
- {file_2} - {change_type} - {line_count} lines
- {file_3} - {change_type} - {line_count} lines

Total: {file_count} files, {total_lines} lines changed

Step 5: Update Story File (1 minute)

Add commit information to story file:

## Implementation Commit
**Developer:** {dev_name} | **Date:** {YYYY-MM-DD} | **Commit:** {commit_hash}

### Commit Summary
- **Message:** {commit_title}
- **Files Changed:** {file_count}
- **Lines Changed:** {total_lines}
- **Quality Gates:** {pass_count} PASS, {fail_count} FAIL

### PR Context Prepared
- Business summary: COMPLETE
- Technical changes: COMPLETE  
- Learning extraction: COMPLETE
- Validation evidence: COMPLETE
- Ready for PR creation: YES

Success Criteria

  • All implementation changes committed to git
  • Commit message follows structured format with business context
  • PR context document prepared with comprehensive details
  • Story file updated with commit information
  • All quality gates confirmed passing before commit
  • Learning items integrated into PR context

Commit Message Guidelines

  • Title: Concise epic-story identifier with brief summary
  • Body: Structured format with business and technical context
  • Learning: Include learning items count and key insights
  • Quality: Include quality gate results and review metrics
  • Attribution: Standard Claude Code attribution

PR Context Structure

  • Business-First: Lead with business value and impact
  • Technical-Second: Detailed technical changes and architecture
  • Learning-Third: Captured learnings and future actions
  • Evidence-Last: Validation proof and review results

Integration Points

  • Input from: party-mode-learning-review (team consensus)
  • Output to: create-comprehensive-pr (PR generation)
  • Handoff: "Implementation committed. Comprehensive PR context prepared. Ready for PR creation."

LLM Optimization Notes

  • Structured commit messages enable rapid parsing
  • Token limits in PR context prevent information overload
  • Business-first ordering prioritizes stakeholder needs
  • Evidence-based validation provides objective review criteria
  • Comprehensive context reduces PR review time and questions