Agent Skills: Execute Checklist Task

Execute Checklist Task

When to Use

• Before marking a story as complete (Definition of Done)
• During sprint planning to validate completeness
• When validating migration progress (strangler pattern)
• Before retrospectives to ensure thoroughness
• When reviewing draft stories for quality

Quick Start

1. Identify checklist to use (or list available from checklists/)
2. Choose mode: Interactive (step-by-step) or YOLO (all at once)
3. Gather required documents/artifacts
4. Process each checklist item
5. Generate summary report with pass/fail status

Purpose

Systematically validate documents, stories, or processes against defined checklists to ensure quality and completeness.

Available Checklists

Check ./checklists/ folder for:

• story-dod-checklist.md - Story definition of done
• story-draft-checklist.md - Story draft validation
• retrospective-checklist.md - Retrospective effectiveness

Instructions

1. Initial Assessment

Determine Checklist:

• If user provides checklist name, fuzzy match it
• If multiple matches, ask for clarification
• If none specified, list available options
• Load checklist from ./checklists/

Choose Processing Mode:

• Interactive: Section by section with discussion
• YOLO (recommended): Process all at once, summary at end

2. Document Gathering

Each checklist specifies required artifacts:

• Stories may need: PRD, architecture, epics
• Sprint planning needs: backlog, velocity data
• Retrospective needs: sprint metrics, burndown

Locate documents per checklist requirements.

3. Checklist Processing

Interactive Mode: For each section:

• Review all items per section instructions
• Check against documentation/artifacts
• Present findings (warnings, errors, N/A items)
• Get confirmation before proceeding
• Halt on major issues if needed

YOLO Mode:

• Process all sections sequentially
• Create comprehensive report
• Present complete analysis
• Highlight critical issues

4. Validation Approach

For each item:

validation:
  item: "Checklist item text"
  status: "[x] Complete | [ ] Missing | [!] Warning | [N/A] Not Applicable"
  finding: "What was found or missing"
  severity: "Critical | Major | Minor | Info"
  recommendation: "Specific action if needed"
  evidence: "Link to doc/section supporting status"

5. Scoring (If Applicable)

Some checklists include scoring:

scoring:
  section_scores:
    requirements: 8/10
    design: 7/10
    testing: 9/10
  total_score: 24/30 (80%)
  threshold: 75% (PASS/FAIL)
  confidence: "High | Medium | Low"

6. Generate Report

Report Structure:

# Checklist Validation Report

## Summary
- Checklist: {name}
- Document: {what was validated}
- Date: {timestamp}
- Mode: {Interactive|YOLO}
- Overall Status: {PASS|FAIL|CONDITIONAL}

## Critical Findings
[List any blockers or critical issues]

## Section Results

### Section 1: {Name}
- Items Checked: X
- Passed: Y
- Warnings: Z
- Not Applicable: W

Key Findings:
- [Finding 1]
- [Finding 2]

### Section 2: {Name}
[Continue for all sections]

## Recommendations

### Immediate Actions
1. {Critical fix needed}
2. {Blocker to resolve}

### Improvements
1. {Enhancement suggestion}
2. {Process improvement}

## Metrics
- Total Items: X
- Compliance Rate: Y%
- Critical Issues: Z
- Estimated Fix Time: {hours}

## Next Steps
1. Address critical issues
2. Review with team
3. Update documentation
4. Re-run validation

7. PSP/TSP Integration

For story checklists, include:

• Estimation accuracy check
• Size category validation
• Time tracking fields present
• Historical proxy appropriateness

For sprint checklists:

• Team role assignments complete
• Velocity goals defined
• Quality metrics established
• Measurement plan in place

8. Follow-Up Actions

Based on validation results:

If PASS:

• Document approval
• Proceed with next phase
• Archive validation report

If CONDITIONAL:

• List specific conditions
• Set review checkpoint
• Assign resolution owners

If FAIL:

• Stop progression
• Document blockers
• Create fix tasks
• Schedule re-validation

Success Criteria

• [ ] Checklist fully processed
• [ ] All items validated with evidence
• [ ] Report generated with findings
• [ ] Recommendations documented
• [ ] Next steps clear
• [ ] Stakeholders informed

Output

• Validation report (markdown)
• Updated checklist with statuses
• Action items if needed
• Metrics for tracking