Analyze

Cross-Artifact Consistency Analyzer

Validate consistency across feature artifacts before implementation.

SpecKit Workflow

This skill is Step 5 of 6 in the Relentless workflow:

specify → plan → tasks → convert → analyze → implement

Prerequisites (ALL must exist):

• spec.md - Feature specification
• plan.md - Technical implementation plan
• tasks.md - User stories and tasks
• prd.json - Converted PRD with routing metadata (run convert first)
• checklist.md - Quality validation checklist (recommended)

Purpose:

• Validate consistency across ALL artifacts
• Catch gaps before implementation begins
• Ensure TDD and routing compliance
• Verify quality gates are defined

The Job

1. Read all feature artifacts
2. Check for consistency issues
3. Validate TDD compliance
4. Validate routing metadata
5. Validate quality gates
6. Validate against constitution
7. Generate analysis report

Files to Analyze

Required:

• relentless/constitution.md
• relentless/features/NNN-feature/spec.md
• relentless/features/NNN-feature/plan.md
• relentless/features/NNN-feature/tasks.md
• relentless/features/NNN-feature/prd.json
• relentless/features/NNN-feature/checklist.md

Optional:

• relentless/features/NNN-feature/progress.txt

Validation Checks

1. Completeness

• All functional requirements from spec have tasks
• All tasks reference spec requirements
• Plan addresses all spec requirements
• Checklist covers all tasks

2. TDD Compliance Validation (MANDATORY)

Verify that:

• Every story in tasks.md has test acceptance criteria
• Test specifications exist in plan.md
• Checklist includes test validation items
• No stories marked "tests optional"
• Given/When/Then format used in acceptance criteria

Flag as CRITICAL if TDD requirements are missing.

3. Routing Metadata Validation (MANDATORY)

Verify that prd.json contains for EVERY story:

• routing.complexity - simple/medium/complex/expert
• routing.harness - claude/gemini/codex/opencode/amp/droid
• routing.model - specific model name
• routing.mode - free/cheap/good/genius
• routing.estimatedCost - cost in USD

Also verify:

• Routing preference in spec.md matches prd.json routing mode
• Complexity classifications are reasonable for story scope

Flag as CRITICAL if routing metadata is missing.

4. Quality Gates Validation (MANDATORY)

Verify that ALL THREE quality gates are present in acceptance criteria:

• Typecheck requirement (bun run typecheck passes with 0 errors)
• Lint requirement (bun run lint passes with 0 errors AND 0 warnings)
• Test requirement (bun test passes)

Flag as CRITICAL if any quality gate is missing.

5. Constitution Compliance

• Plan follows MUST rules
• Tasks include required quality checks
• Testing requirements met
• Security requirements addressed
• No any types planned

6. Dependency Consistency

• Task dependencies are valid (no circular refs)
• Referenced dependencies exist
• Dependency order makes sense

7. Scope Consistency

• Plan doesn't include out-of-scope items
• Tasks match plan scope
• No scope creep in checklist

8. Data Model Consistency

• Entities mentioned in spec are in plan
• Plan data models match spec requirements
• Tasks create required entities

9. API Consistency

• Endpoints in plan match spec scenarios
• Tasks implement all planned endpoints
• Checklist validates all endpoints

10. Testing Coverage

• Each task has test acceptance criteria
• Checklist includes test validation
• Test types match constitution requirements

Issue Severity

CRITICAL: Blocks implementation - MUST fix before proceeding

• Constitution MUST rule violations
• Missing TDD requirements
• Missing routing metadata in prd.json
• Missing quality gate requirements
• Circular dependencies
• Missing required artifacts

WARNING: Should fix before implementation

• Incomplete coverage of requirements
• Ambiguous acceptance criteria
• Missing test criteria for some stories
• Routing preference mismatch (spec vs prd.json)

INFO: Nice to have

• Style inconsistencies
• Minor gaps in checklist
• Optimization opportunities

Report Format

Use the template at templates/analysis-report.md:

# Consistency Analysis Report

**Feature:** [feature-name]
**Date:** [date]
**Status:** [PASS | FAIL | WARNINGS]

## Summary

| Category | Status | Issues |
|----------|--------|--------|
| Completeness | ✅/❌ | [count] |
| TDD Compliance | ✅/❌ | [count] |
| Routing Validation | ✅/❌ | [count] |
| Quality Gates | ✅/❌ | [count] |
| Constitution | ✅/❌ | [count] |
| Dependencies | ✅/❌ | [count] |

## Critical Issues

### CRIT-001: [Issue Title]
**Location:** [file]
**Details:** [description]
**Fix:** [required action]

## Recommendations

1. [Action item 1]
2. [Action item 2]

**Next Step:** /relentless.implement (if PASS) or fix issues (if FAIL)

Output

1. Save analysis report to relentless/features/NNN-feature/analysis-report.md
2. Update progress.txt with analysis results
3. Report:
   • Overall status: PASS/FAIL/WARNINGS
   • Critical issues count
   • TDD compliance: PASS/FAIL
   • Routing validation: PASS/FAIL
   • Quality gates: PASS/FAIL
   • Next step: /relentless.implement or fix issues

Notes

• Run before starting implementation
• Constitution compliance is non-negotiable
• TDD compliance is non-negotiable
• Routing metadata is required for all stories
• Quality gates (typecheck, lint, test) are mandatory
• Fix critical issues immediately
• Warnings can sometimes be deferred
• Re-run after making fixes