refactor: update epic retrospective task and configuration

- Renamed retrospective file pattern in core-config.yml to align with new naming conventions.
- Removed outdated references to the epic-retrospective task in QA agent documentation.
- Enhanced the epic retrospective task documentation to focus on AI agent development processes and collaboration.
- Updated the retrospective template to reflect changes in analysis focus and required inputs.
- Adjusted user guide to include retrospectives as part of the Scrum Master's responsibilities.

bmad-core/agents/qa.md

@@ -43,9 +43,8 @@ commands:  # All commands require * prefix when used (e.g., *help)
 dependencies:
   tasks:
     - review-story
-    - epic-retrospective
   data:
     - technical-preferences
   utils:
     - template-format
-```
+```

bmad-core/agents/sm.md

@@ -34,6 +34,7 @@ commands:  # All commands require * prefix when used (e.g., *help)
   - help: Show numbered list of the following commands to allow selection
   - chat-mode: Conversational mode with advanced-elicitation for advice
   - create|draft: Execute create-next-story
+  - retrospective: Execute epic-retrospective task
   - pivot: Execute `correct-course` task
   - checklist {checklist}: Show numbered list of checklists, execute selection
   - exit: Say goodbye as the Scrum Master, and then abandon inhabiting this persona
@@ -42,10 +43,11 @@ dependencies:
     - create-next-story
     - execute-checklist
     - course-correct
+    - epic-retrospective
   templates:
     - story-tmpl
+    - epic-retrospective-tmpl
   checklists:
     - story-draft-checklist
   utils:
-    - template-format
+    - template-format
-```

bmad-core/core-config.yml

@@ -14,7 +14,7 @@ architecture:
 retrospectives:
   enabled: true
   retrospectiveLocation: docs/retrospectives
-  retrospectiveFilePattern: qa-epic-retrospective-*.md
+  retrospectiveFilePattern: epic-retrospective-*.md
   depth: 3  # Number of recent retrospectives to consider
   maxDepth: 5  # Maximum allowed depth for safety
 customTechnicalDocuments: null

bmad-core/tasks/epic-retrospective.md

@@ -1,30 +1,30 @@
-# QA Epic Retrospective Task
+# Epic Retrospective Task
 
 ## Purpose
 
-Conduct a comprehensive retrospective analysis of completed epics from a quality assurance standpoint. This task focuses on evaluating the entire development and testing process to identify successes, failures, and areas for improvement in product quality and testing strategy. The resulting retrospective document serves as a learning tool to guide future development cycles and enhance overall quality processes.
+Facilitate a comprehensive retrospective analysis of completed epics from a team and process improvement standpoint. This task focuses on evaluating the AI agent development process, chat interactions, and commit history to identify successes, challenges, and areas for improvement. As the Scrum Master, you facilitate this retrospective to gather insights from development experience and guide future epics toward better AI agent collaboration and more effective development workflows.
 
 ## When to Use This Task
 
 **Use this task when:**
 
-- An epic has been completed and deployed
+- An epic has been completed with all stories reaching Definition of Done
-- All associated stories have reached their Definition of Done
+- AI agent development sessions have concluded for the epic
-- QA wants to capture lessons learned for process improvement
+- There is sufficient chat history and commit data to analyze
-- Team is ready to invest time in reflective analysis
+- The team is ready to invest time in process improvement analysis
 
 **Use alternative retrospective approaches when:**
 
-- Epic is still in progress (use review-story instead)
+- Epic is still in progress (use review-story instead for individual story feedback)
-- Immediate hotfixes are needed (focus on incident post-mortems)
+- Immediate bug fixes are needed (focus on incident analysis)
-- Limited QA involvement in the epic (consider development-focused retrospectives)
+- Epic was abandoned or significantly changed mid-development (consider lessons learned session)
-- Epic was abandoned or significantly changed mid-development
+- Insufficient development history exists for meaningful analysis
 
 ## Instructions
 
 ### 1. Data Collection and Preparation (Required)
 
-Before beginning the analysis, gather all relevant documentation and data from the completed epic:
+As the Scrum Master, gather comprehensive development artifacts and AI agent interaction data:
 
 **Epic Documentation:**
 
@@ -33,68 +33,79 @@ Before beginning the analysis, gather all relevant documentation and data from t
 - [ ] Definition of Done criteria for each story
 - [ ] Acceptance criteria and business requirements
 
-**Testing Documentation:**
+**AI Agent Development Data:**
 
-- [ ] Test plans and test case documentation
+- [ ] Chat transcripts from agent interactions during epic development
-- [ ] Test execution reports and results
+- [ ] Agent handoff quality between SM → Dev → QA → PO workflows
-- [ ] Bug reports filed during the epic development
+- [ ] Story preparation effectiveness and clarity
-- [ ] Performance test results (if applicable)
+- [ ] Agent instruction adherence and task completion patterns
-- [ ] User acceptance testing feedback
+- [ ] Template usage and effectiveness during development
 
-**Development Artifacts:**
+**Development Process Data:**
 
-- [ ] Git commit history for the epic's duration
+- [ ] Git commit history and development patterns throughout the epic
-- [ ] Final codebase snapshot for the epic
+- [ ] Story completion rates and development velocity
+- [ ] Code review outcomes and refactoring patterns
+- [ ] Testing implementation and coverage achieved
+- [ ] Debug log entries and issue resolution patterns
 
-### 2. Quality Analysis
+**Technical Artifacts:**
 
-Conduct a thorough analysis of the collected data from multiple QA perspectives:
+- [ ] Final codebase state for the epic
+- [ ] Test coverage reports and quality metrics
+- [ ] Build and validation outcomes
+- [ ] Architecture adherence and technical debt introduction
 
-#### Test Coverage and Effectiveness Analysis
+### 2. AI Agent Development Process Analysis
 
-**Test Plan Review:**
+Conduct a thorough analysis of the AI-assisted development workflow:
 
-- [ ] Analyze test plan coverage against requirements
+#### Agent Collaboration and Workflow Analysis
-- [ ] Identify gaps between planned and executed tests
-- [ ] Evaluate test case quality and completeness
-- [ ] Review automation vs manual testing balance
 
-**Bug Analysis:**
+**BMAD Agent Effectiveness:**
 
-- [ ] Categorize bugs by severity and type
+- [ ] Evaluate story preparation quality from SM agent
-- [ ] Analyze bug introduction points in development cycle
+- [ ] Assess development task execution by Dev agent
-- [ ] Review bug fix effectiveness and regression occurrences
+- [ ] Review code quality improvements from QA agent
-- [ ] Identify patterns in bug types or areas
+- [ ] Analyze handoff clarity between different agents
 
-#### Documentation and Requirements Analysis
+**Agent Instruction Adherence:**
 
-**Requirements Validation:**
+- [ ] Review how well agents followed story instructions
+- [ ] Assess template usage and consistency
+- [ ] Identify instruction ambiguities that caused confusion
+- [ ] Evaluate agent decision-making quality
 
-- [ ] Verify final product meets acceptance criteria
+#### Development Process Analysis
-- [ ] Identify requirements that were unclear or changed
-- [ ] Assess story Definition of Done effectiveness
-- [ ] Review business value delivery against original goals
 
-**Documentation Quality:**
+**Story Execution Quality:**
 
-- [ ] Evaluate test documentation completeness
+- [ ] Analyze story completion patterns and blockers
-- [ ] Review user documentation accuracy
+- [ ] Review task breakdown effectiveness
-- [ ] Assess technical documentation for QA processes
+- [ ] Assess testing integration during development
+- [ ] Evaluate Definition of Done achievement
 
-#### Codebase and Development Process Analysis
+**Code Quality and Patterns:**
 
-**Code Quality Review:**
+- [ ] Review commit history for development patterns
+- [ ] Assess code quality and architectural consistency
+- [ ] Analyze refactoring effectiveness
+- [ ] Evaluate technical debt management
 
-- [ ] Analyze commit history for quality trends
+#### Chat and Communication Analysis
-- [ ] Review code review effectiveness from QA perspective
-- [ ] Identify areas with high bug density
-- [ ] Assess testability of delivered code
 
-**Process Effectiveness:**
+**Agent Communication Effectiveness:**
 
-- [ ] Evaluate QA involvement throughout development
+- [ ] Review chat clarity and instruction quality
-- [ ] Review testing environment stability and effectiveness
+- [ ] Assess agent response accuracy and helpfulness
-- [ ] Analyze deployment and release process quality
+- [ ] Identify communication breakdowns or misunderstandings
+- [ ] Evaluate iterative improvement patterns
+
+**Development Flow:**
+
+- [ ] Analyze development velocity and momentum
+- [ ] Review blocker identification and resolution
+- [ ] Assess feedback incorporation effectiveness
 
 ### 3. Retrospective Document Generation
 
@@ -102,32 +113,33 @@ Use the epic-retrospective-tmpl.md template to create the comprehensive retrospe
 
 #### What Went Well Section
 
-Focus on QA successes and effective practices:
+Focus on effective AI agent development practices:
 
-- **Testing Strategies:** Note effective testing approaches, tools, or methodologies
+- **Agent Coordination:** Note effective handoffs and collaboration patterns
-- **Collaboration:** Highlight good communication between QA and development teams
+- **Story Quality:** Highlight well-prepared stories that led to smooth development
-- **Quality Metrics:** Document areas with low bug counts or high test coverage
+- **Development Flow:** Document efficient development patterns and velocity
-- **Process Improvements:** Recognize successful process changes implemented during the epic
+- **Code Quality:** Celebrate effective testing, refactoring, and architectural decisions
+- **Template Usage:** Recognize effective use of BMAD templates and processes
 
 #### What Could Be Improved Section
 
-Identify specific areas needing enhancement:
+Identify specific areas needing enhancement in AI agent development:
 
-- **Test Coverage Gaps:** Document areas with insufficient testing
+- **Agent Instructions:** Note unclear or ambiguous instructions that caused issues
-- **Bug Prevention:** Identify recurring bug types or root causes
+- **Story Preparation:** Identify gaps in story technical guidance or context
-- **Requirements Issues:** Note unclear acceptance criteria or changing requirements
+- **Development Process:** Document workflow inefficiencies or agent confusion
-- **Environment Issues:** Document testing environment problems or limitations
+- **Template Effectiveness:** Note template limitations or unclear sections
-- **Communication:** Identify breakdowns in QA/Dev collaboration
+- **Communication Patterns:** Identify areas where agent communication could improve
 
 #### Actionable Recommendations Section
 
-Propose concrete, implementable improvements:
+Propose concrete improvements for future AI agent development:
 
-- **Process Changes:** Specific modifications to QA processes
+- **Story Template Updates:** Specific improvements to story preparation templates
-- **Tool Improvements:** New tools or better use of existing tools
+- **Agent Instruction Clarity:** Better instruction patterns for agents
-- **Training Needs:** Skills development for team members
+- **Workflow Optimization:** Process improvements for agent handoffs
-- **Documentation Updates:** Improvements to templates or standards
+- **Template Enhancements:** Updates to BMAD templates and checklists
-- **Automation Opportunities:** Areas where manual testing could be automated
+- **Development Practice:** Better patterns for AI-assisted development
 
 ### 4. Validation and Review
 
@@ -135,54 +147,55 @@ Before finalizing the retrospective, ensure comprehensive coverage:
 
 **Content Validation:**
 
-- [ ] All major quality issues identified and analyzed
+- [ ] All major development process issues identified and analyzed
-- [ ] Recommendations are specific and actionable
+- [ ] Recommendations are specific, actionable, and focused on AI agent workflows
-- [ ] Both positive and negative aspects are balanced
+- [ ] Both successful and challenging aspects are balanced
-- [ ] Data supports all conclusions and recommendations
+- [ ] Analysis is supported by chat history and commit evidence
 
-**Stakeholder Review:**
+**Process Review:**
 
-- [ ] Development team feedback incorporated
+- [ ] Agent workflow effectiveness assessed
-- [ ] Product Owner perspective considered
+- [ ] Story preparation quality evaluated
-- [ ] Scrum Master/Project Manager input included
+- [ ] Template and instruction clarity reviewed
-- [ ] Technical Architecture feedback if relevant
+- [ ] Development velocity and quality patterns analyzed
 
 **Follow-up Planning:**
 
-- [ ] Recommendations prioritized by impact and effort
+- [ ] Recommendations prioritized by impact on future agent development
-- [ ] Owners assigned for implementing recommendations
+- [ ] Template and process updates identified
-- [ ] Timeline established for process improvements
+- [ ] Agent instruction improvements specified
 - [ ] Success metrics defined for recommended changes
 
 ## Output
 
-A single markdown file named `qa-epic-retrospective-<epic_name>.md` saved in the `docs/retrospectives/` directory.
+A single markdown file named `epic-retrospective-<epic_name>.md` saved in the `docs/retrospectives/` directory.
 
 **File Structure Requirements:**
 
 - Use the epic-retrospective-tmpl.md template
-- Include all sections with substantive content
+- Include all sections with focus on AI agent development process
-- Provide specific examples rather than generic statements
+- Provide specific examples from chat history and commit patterns
-- Include metrics and data where available
+- Include metrics and evidence from development artifacts
-- Ensure recommendations are actionable with clear next steps
+- Ensure recommendations improve future AI agent collaboration
 
 ## Success Criteria
 
 The epic retrospective is successful when:
 
-1. **Comprehensive Analysis:** All aspects of quality and testing are thoroughly reviewed
+1. **Comprehensive Process Analysis:** All aspects of AI agent development workflow are reviewed
-2. **Data-Driven Insights:** Conclusions are supported by concrete evidence from the epic
+2. **Evidence-Based Insights:** Conclusions are supported by chat history and commit data
-3. **Actionable Recommendations:** Proposed improvements are specific, measurable, and implementable
+3. **Actionable Recommendations:** Proposed improvements are specific to AI agent development
-4. **Process Learning:** Document provides clear guidance for future epic development
+4. **Process Learning:** Document provides clear guidance for future AI-assisted epic development
-5. **Stakeholder Value:** Insights are valuable to development team, product management, and QA processes
+5. **Template Improvement:** Identifies concrete ways to enhance BMAD templates and processes
-6. **Continuous Improvement:** Recommendations address root causes, not just symptoms
+6. **Agent Workflow Optimization:** Recommendations improve agent collaboration and effectiveness
 
 ## Important Notes
 
-- **Timing Matters:** Conduct retrospectives while the epic experience is still fresh, but allow time for post-deployment observation
+- **Focus on Development Process:** Analyze AI agent interactions, not real-world deployment outcomes
-- **Focus on Learning:** Emphasize learning and improvement over blame or criticism
+- **Chat History Analysis:** Use actual agent conversations to identify patterns and issues
-- **Data Over Opinion:** Base conclusions on concrete evidence from testing and development artifacts
+- **Commit Pattern Review:** Analyze development velocity and code quality trends
-- **Actionable Outcomes:** Ensure every recommendation has a clear path to implementation
+- **Template Effectiveness:** Evaluate how well BMAD templates guided development
-- **Follow Through:** Plan to review retrospective recommendations in future epics to ensure continuous improvement
+- **Agent Instruction Quality:** Focus on clarity and effectiveness of agent instructions
+- **Iterative Improvement:** Recommendations should enhance future AI agent development cycles
 
-**Agent Responsibility:** The QA agent conducting this retrospective should maintain objectivity while advocating for quality improvements that benefit the entire development process.
+**Agent Responsibility:** The Scrum Master conducting this retrospective should focus on improving the AI-assisted development process, agent collaboration patterns, and template effectiveness for future epics.

bmad-core/templates/epic-retrospective-tmpl.md

@@ -1,25 +1,27 @@
-# Epic Retrospective: {{epic_name}}
+# Epic Retrospective: epic-retrospective-{{epic_name}}
 
-[[LLM: The default path and filename unless specified is docs/retrospectives/qa-epic-retrospective-{{epic_name}}.md]]
+[[LLM: The default path and filename unless specified is docs/retrospectives/epic-retrospective-{{epic_name}}.md]]
 
-[[LLM: This template guides comprehensive QA-focused retrospective analysis of completed epics. Before beginning:
+[[LLM: This template guides comprehensive team retrospective analysis of completed epics facilitated by the Scrum Master. Before beginning:
 
-1. **Verify Epic Completion**: Confirm the epic has been fully deployed and sufficient time has passed for real-world observation (recommended: 1-2 weeks post-deployment)
+1. **Verify Epic Completion**: Confirm the epic has been fully completed with all stories reaching Definition of Done
 
-2. **REQUIRED INPUTS**: Ensure access to all necessary documentation and data:
+2. **REQUIRED INPUTS**: Ensure access to all necessary development artifacts and AI agent interaction data:
    - Completed epic document with original goals and scope
    - All story documents with Definition of Done criteria
-   - Test plans, test cases, and execution reports
+   - Chat transcripts from AI agent interactions during epic development
-   - Bug reports and issue tracking data
+   - Git commit history and development patterns throughout the epic
-   - Git commit history for the epic duration
+   - Story completion rates and development velocity metrics
-   - Final codebase snapshot
+   - Agent handoff quality and workflow effectiveness data
-   - Performance metrics and monitoring data
+   - Template usage and instruction adherence patterns
 
-3. **DEEP ANALYSIS MANDATE**: You MUST conduct thorough analysis of the actual development and testing process. Every insight must be based on concrete evidence from the epic's execution, not assumptions.
+3. **AI DEVELOPMENT PROCESS FOCUS**: As Scrum Master, you MUST analyze the AI-assisted development workflow. Every insight must be based on concrete evidence from chat history, commit patterns, and agent interactions, not assumptions.
 
-4. **CONTINUOUS VALIDATION**: Throughout this process, validate your understanding with stakeholders. For every conclusion, confirm: "Based on my analysis of the epic data, I found [specific evidence] which indicates [conclusion]. Does this align with the team's experience?"
+4. **DEVELOPMENT WORKFLOW ANALYSIS**: Focus on AI agent collaboration patterns, story preparation effectiveness, and development process optimization.
 
-If any required inputs are missing, request them before proceeding.
+5. **COMPREHENSIVE VALIDATION**: Throughout this process, validate understanding based on development artifacts. For every conclusion, confirm: "Based on chat history and commit evidence, I found [specific evidence] which indicates [conclusion]. This pattern suggests [improvement opportunity]."
+
+If any required development artifacts or agent interaction data are missing, gather them before proceeding.
 
 After presenting this introduction, apply `tasks#advanced-elicitation` protocol]]
 

docs/user-guide.md

@@ -112,16 +112,16 @@ The reason #2 and #3 are optional is because now BMAD V4 makes sharding optional
 
 ### Core Development Team
 
-| Agent       | Role               | Primary Functions                              | When to Use                                       |
+| Agent       | Role               | Primary Functions                               | When to Use                                       |
-| ----------- | ------------------ | ---------------------------------------------- | ------------------------------------------------- |
+| ----------- | ------------------ | ----------------------------------------------- | ------------------------------------------------- |
-| `analyst`   | Business Analyst   | Market research, requirements gathering        | Project planning, competitive analysis            |
+| `analyst`   | Business Analyst   | Market research, requirements gathering         | Project planning, competitive analysis            |
-| `pm`        | Product Manager    | PRD creation, feature prioritization           | Strategic planning, roadmaps                      |
+| `pm`        | Product Manager    | PRD creation, feature prioritization            | Strategic planning, roadmaps                      |
-| `architect` | Solution Architect | System design, technical architecture          | Complex systems, scalability planning             |
+| `architect` | Solution Architect | System design, technical architecture           | Complex systems, scalability planning             |
-| `dev`       | Developer          | Sequential task execution, testing, validation | Story implementation with test-driven development |
+| `dev`       | Developer          | Sequential task execution, testing, validation  | Story implementation with test-driven development |
-| `qa`        | QA Specialist      | Code review, refactoring, test validation      | Senior developer review via `review-story` task   |
+| `qa`        | QA Specialist      | Code review, refactoring, test validation       | Senior developer review via `review-story` task   |
-| `ux-expert` | UX Designer        | UI/UX design, prototypes                       | User experience, interface design                 |
+| `ux-expert` | UX Designer        | UI/UX design, prototypes                        | User experience, interface design                 |
-| `po`        | Product Owner      | Backlog management, story validation           | Story refinement, acceptance criteria             |
+| `po`        | Product Owner      | Backlog management, story validation            | Story refinement, acceptance criteria             |
-| `sm`        | Scrum Master       | Sprint planning, story creation                | Project management, workflow                      |
+| `sm`        | Scrum Master       | Sprint planning, story creation, retrospectives | Project management, workflow                      |
 
 ### Meta Agents
 
@@ -1163,7 +1163,7 @@ coreProjectLocation:
   retrospectives: # Retrospective integration settings
     enabled: true # Enable retrospective insights in story creation
     retrospectiveLocation: docs/retrospectives # Where retrospective files are stored
-    retrospectiveFilePattern: qa-epic-retrospective-*.md # Pattern for retrospective files
+    retrospectiveFilePattern: epic-retrospective-*.md # Pattern for retrospective files
     depth: 3 # Number of recent retrospectives to consider (default: 3)
     maxDepth: 5 # Maximum allowed depth for safety
 
@@ -1254,13 +1254,13 @@ This ensures the dev agent always has critical context without needing to search
 
 ##### Retrospective Configuration
 
-BMAD V4 introduces intelligent retrospective integration that helps teams learn from past epics and apply those insights to new story creation:
+BMAD V4 introduces intelligent retrospective integration that helps teams learn from past AI agent development experiences and apply those insights to improve future story creation and development workflows:
 
 ```yaml
 retrospectives:
   enabled: true # Enable/disable retrospective integration
   retrospectiveLocation: docs/retrospectives # Where retrospective files are stored
-  retrospectiveFilePattern: qa-epic-retrospective-*.md # Pattern for retrospective files
+  retrospectiveFilePattern: epic-retrospective-*.md # Pattern for retrospective files
   depth: 3 # Number of recent retrospectives to consider
   maxDepth: 5 # Maximum allowed depth for safety
 ```