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/validate-consolidated-fixes.md

14 KiB
Raw Blame History

validate-consolidated-fixes

Agent: architect
Type: BatchTask

Purpose

Validate that consolidated fixes have been properly implemented by examining story documentation and using Playwright MCP for UX changes, serving as the single validation gate for Round 2+ iterations.

Context

This task provides efficient architect-only validation after comprehensive Round 1 reviews:

  • Validates fixes against original consolidated feedback
  • Uses story documentation as primary evidence source
  • Employs Playwright MCP for UX validation when needed
  • Provides final approval or requests additional fixes
  • Eliminates need for full re-review cycles

Inputs

Required

  • story_file (string): Path to story file with implementation details
  • original_feedback (object): Original consolidated feedback for comparison
  • implemented_fixes (object): Summary of changes implemented by dev

Outputs

  • validation_status (string): "APPROVED" | "NEEDS_FIXES" | "BLOCKED"
  • validation_results (object): Detailed validation findings
  • additional_feedback (string): Specific guidance for any remaining issues
  • story_file (string): Updated story file with validation results

Instructions

Pre-Validation Analysis

  1. Review implementation documentation

    • Read story file implementation section thoroughly
    • Compare implemented fixes against original consolidated feedback
    • Identify any REQUIRED-FOR-COMPLETION items that were not addressed
    • Note any technical decisions or changes made during implementation

  2. Assess validation approach needed

    • Identify which fixes require technical code review
    • Determine which changes need UX validation via Playwright
    • Note any business logic changes requiring functional testing
    • Plan validation sequence based on dependencies

Validation Process

  1. Validate technical fixes

    Architecture fixes validation:

    • Review code changes described in story documentation
    • Verify security improvements implemented correctly
    • Check performance optimizations are reasonable
    • Confirm technical debt addressed appropriately

    Quality fixes validation:

    • Verify test coverage improvements documented
    • Check that quality gates are passing
    • Confirm code quality standards maintained
    • Validate error handling additions

  2. Validate business fixes

    Acceptance criteria validation:

    • Confirm AC adjustments align with business requirements
    • Verify business rule corrections documented properly
    • Check user journey improvements are logical
    • Validate data validation enhancements

    Epic alignment validation:

    • Ensure changes maintain epic scope and objectives
    • Verify business value delivery still intact
    • Check that project phase boundaries respected

  3. Validate UX fixes using project-appropriate testing tools

    When UX validation needed:

    • User interface changes described in story (web, desktop, mobile, CLI, etc.)
    • User interaction flow modifications
    • Accessibility improvements requiring testing
    • Design consistency updates

    Comprehensive UI Testing Protocol:

    Phase 1: Environment Setup

    • Launch UI testing tools as specified in project documentation (check README.md or test configuration)
    • Use clean testing environment appropriate for project type (browser, desktop app, mobile simulator, CLI, etc.)
    • Clear relevant caches and state according to project architecture
    • Set appropriate viewport/window size for project target platform
    • Configure testing environment for debugging (enable logging as per project standards)

    Phase 2: Pre-Testing Validation

    • Navigate to application entry point (URL, app launch, CLI command, etc.)
    • Verify application loads/starts without errors (check relevant logs)
    • Take baseline screenshot/capture of unaffected areas for comparison
    • Document initial application state and version

    Phase 3: Feature-Specific Testing

    • Navigate systematically to each UI area mentioned in story fixes
    • For each changed component/page:
      • Capture screenshot BEFORE interacting
      • Test all documented user interactions (clicks, form submissions, navigation)
      • Verify visual changes match story implementation descriptions
      • Test error states and edge cases if applicable
      • Capture screenshot AFTER each significant interaction
      • Validate loading states and transitions work correctly

    Phase 4: Accessibility & Responsive Testing (if applicable to project type)

    • Test keyboard/alternative navigation for new/changed interactive elements
    • Verify accessibility features if improvements documented (ARIA, screen reader compatibility, etc.)
    • Test responsive behavior according to project target platforms (mobile, tablet, desktop, multiple screen sizes)
    • Validate contrast and readability for visual changes according to project standards

    Phase 5: Cross-Platform Compatibility (if critical changes)

    • Repeat core tests across project target platforms (different browsers, OS versions, device types, etc.)
    • Document any platform-specific issues discovered
    • Capture comparative evidence across platforms for visual/behavioral changes

    Phase 6: Evidence Documentation and Cleanup

    • Save all screenshots/captures to temporary validation directory with descriptive filenames (feature_state_timestamp.png)
    • Record any errors or warnings encountered in relevant logs
    • Document specific testing commands/tools used for reproducibility
    • Create testing summary with pass/fail status for each tested component
    • Note: All testing artifacts are temporary and will be cleaned up after validation completion

    Testing Session Management:

    • Maintain consistent testing context throughout validation for consistency
    • Reset application state between major test sections to ensure clean state
    • Restart testing environment if session becomes unstable
    • Document testing tool versions and configuration used (refer to project documentation)
    • Clean up testing sessions and temporary files after validation

    File Management:

    • All screenshots and evidence saved to temporary validation workspace
    • Artifacts automatically cleaned up after validation completion
    • Only validation results and decisions persisted in story documentation
    • No permanent files created during browser testing process

Validation Decision Making

  1. Assess overall fix quality

    APPROVED criteria:

    • All REQUIRED-FOR-COMPLETION items addressed satisfactorily
    • All QUALITY-STANDARD items addressed per project requirements
    • Quality gates passing
    • UX changes validated via browser MCP testing (if applicable)
    • No new issues introduced
    • Documentation clear and complete

    NEEDS_FIXES criteria:

    • Some REQUIRED-FOR-COMPLETION or QUALITY-STANDARD items incomplete or incorrect
    • Quality gates failing
    • UX changes not working as expected
    • Minor issues that can be corrected quickly

    BLOCKED criteria:

    • Major technical blockers preventing completion
    • Fundamental misunderstanding of requirements
    • Scope changes required beyond current story
    • Environment or infrastructure issues

  2. Document validation results

    Update story file with validation findings:

    ## Round 2+ Validation Results

**Validation Date**: [Current date]
**Validation Status**: [APPROVED/NEEDS_FIXES/BLOCKED]

### Architecture Fixes Validation
- [Fix 1]: ✅ VALIDATED / ❌ NEEDS_WORK / ⚠️ CONCERNS
- [Fix 2]: ✅ VALIDATED / ❌ NEEDS_WORK / ⚠️ CONCERNS

### Business Fixes Validation  
- [Fix 1]: ✅ VALIDATED / ❌ NEEDS_WORK / ⚠️ CONCERNS
- [Fix 2]: ✅ VALIDATED / ❌ NEEDS_WORK / ⚠️ CONCERNS

### Quality Fixes Validation
- [Fix 1]: ✅ VALIDATED / ❌ NEEDS_WORK / ⚠️ CONCERNS
- [Fix 2]: ✅ VALIDATED / ❌ NEEDS_WORK / ⚠️ CONCERNS

### UX Fixes Validation (Browser MCP Testing)
**Browser MCP Tool Used:** [Playwright MCP / Puppeteer MCP / Other Browser MCP]
**Testing Session ID:** [Unique identifier for reproducibility]
**Test Environment:** [URL, version, browser details]

**Component-Level Results:**
- [Component 1]: ✅ VALIDATED / ❌ NEEDS_WORK / ⚠️ CONCERNS
  * **Interaction Testing:** [Pass/Fail with specific interactions tested]
  * **Visual Validation:** [Pass/Fail with screenshot evidence]
  * **Accessibility Check:** [Pass/Fail/N/A with specific findings]
  * **Responsive Testing:** [Pass/Fail across viewports]

- [Component 2]: ✅ VALIDATED / ❌ NEEDS_WORK / ⚠️ CONCERNS
  * **Interaction Testing:** [Pass/Fail with specific interactions tested]
  * **Visual Validation:** [Pass/Fail with screenshot evidence]
  * **Accessibility Check:** [Pass/Fail/N/A with specific findings]
  * **Responsive Testing:** [Pass/Fail across viewports]

**Cross-Browser Compatibility:** [Tested/Not Required]
- Chrome: [Pass/Fail/Not Tested] - [Specific findings]
- Firefox: [Pass/Fail/Not Tested] - [Specific findings] 
- Safari: [Pass/Fail/Not Tested] - [Specific findings]

**Evidence Artifacts:**
- Screenshots captured: [Count] files saved with naming convention
- Console errors logged: [Count] with severity levels
- MCP session logs: [Available/Not Available] for debugging

**Overall UX Validation Status:** [PASSED/FAILED/PARTIALLY_PASSED]
**Detailed Findings:** [Comprehensive summary of all UX testing results]

### Additional Feedback (if NEEDS_FIXES)
[Specific, actionable guidance for remaining issues]

### Next Steps
[Clear direction for completion or additional work needed]

Completion Actions

  1. Provide clear next steps

    If APPROVED:

    • Mark story as ready for delivery
    • Document successful completion
    • Note any IMPROVEMENT items for future tracking

    If NEEDS_FIXES:

    • Provide specific, actionable feedback
    • Prioritize remaining issues
    • Set up for another validation cycle
    • Maintain positive momentum

    If BLOCKED:

    • Document blockers clearly
    • Recommend escalation path
    • Suggest scope adjustments if needed
    • Provide technical guidance for resolution

Success Criteria

  • All REQUIRED-FOR-COMPLETION and QUALITY-STANDARD fixes validated against original requirements
  • UX changes comprehensively tested via browser MCP with evidence documentation
  • Browser MCP testing includes interaction validation, visual verification, accessibility checks, and responsive testing
  • Cross-browser compatibility validated for critical changes (Chrome, Firefox, Safari)
  • Quality gates confirmed passing with specific validation evidence
  • Screenshot evidence captured and properly documented for all UI changes
  • Browser MCP session properly managed with clean state testing throughout
  • Clear validation decision made (APPROVED/NEEDS_FIXES/BLOCKED) with comprehensive rationale
  • Story documentation updated with detailed validation results including browser MCP findings
  • Next steps clearly communicated with specific actionable guidance

Failure Conditions

  • Unable to validate fixes due to insufficient documentation
  • Browser MCP testing fails for UX changes
  • Critical regressions discovered during validation
  • Original requirements misunderstood in implementation
  • Technical blockers prevent completion

Error Handling

If documentation is insufficient:

  1. Request specific clarification from dev
  2. Document what additional information is needed
  3. Provide guidance on documentation standards
  4. Validate what can be assessed from available information

If browser MCP testing fails:

  1. Document failure details comprehensively:

    • Specific MCP tool and version used (Playwright MCP, Browser MCP, etc.)
    • Exact failure scenarios with timestamps
    • Browser console errors and MCP session logs
    • Screenshots of failure states if captured

  2. Attempt alternative browser MCP approaches:

    • Try different browser MCP tool if available (switch from Playwright to Browser MCP)
    • Test in different browser engines (Chromium, Firefox, WebKit)
    • Use different viewport sizes to isolate responsive issues
    • Clear browser context completely and retry

  3. Fallback validation methods:

    • Request manual testing documentation from dev with comprehensive screenshots
    • Require video screen recordings of user interactions for complex flows
    • Request specific console log outputs for JavaScript errors
    • Ask for accessibility audit results using browser dev tools

  4. Escalation procedures:

    • Escalate to DevOps if browser MCP infrastructure issues suspected
    • Involve UX Expert for complex accessibility or interaction validation
    • Engage with development team for application-specific testing guidance
    • Consider scope adjustment if UX changes cannot be properly validated via available MCP tools

If validation reveals new issues:

  1. Classify as REQUIRED-FOR-COMPLETION vs IMPROVEMENT
  2. Provide clear guidance for resolution
  3. Update feedback for next implementation cycle
  4. Consider if scope adjustment needed

Notes

  • This task serves as the single validation gate for efficient iterations
  • Focus on validating against original consolidated feedback
  • Use browser MCP tools (Playwright MCP/Puppeteer MCP/similar) for UX changes requiring server interaction
  • Story documentation quality is critical for effective validation
  • Maintain positive, constructive feedback for development team

Integration Points

  • Input from: implement-consolidated-fixes task (dev agent)
  • Output to: Story completion OR additional fix cycles
  • Dependencies: Story file with implementation documentation
  • Tools: Browser MCP tools (Playwright MCP/Puppeteer MCP/similar) for UX validation, project quality gates
  • Escalation: Product Owner for business decisions, DevOps for infrastructure issues