checkpoint

# Checkpoint (Code Review at Task Completion) > **Purpose**: Perform an immediate code review at task/PR completion to catch and fix issues early. > > **Core features**: > - Auto-detect Git Diff > - Auto-extract TASKS.md context > - Auto-invoke `/security-review` > - Hook gate + fix guide --- ## Core Features ### Git Diff Auto-Detection ### Our `/checkpoint` ``` Auto-detect: Git Diff (latest commit) Auto-extract: TASKS.md → match related tasks Hook integration: policy-gate, standards-validator Enhanced security: /security-review auto-invoked AI multi-review: /multi-ai-review optional invocation Result: Pass/Warning/Fail + concrete fix guide ``` --- ## Execution Flow ``` /checkpoint ↓ ┌─────────────────────────────────────────┐ │ Step 1: Auto-detect Git Diff │ │ • git diff HEAD~1 HEAD auto-run │ │ • Extract list of changed files │ └─────────────────────────────────────────┘ ↓ ┌─────────────────────────────────────────┐ │ Step 2: Extract TASKS.md context │ │ • Match changed files ↔ tasks │ │ • Auto-identify related requirements │ └─────────────────────────────────────────┘ ↓ ┌─────────────────────────────────────────┐ │ Step 3: 2-Stage Review │ │ • Stage 1: Spec Compliance │ │ • Stage 2: Code Quality │ └─────────────────────────────────────────┘ ↓ ┌─────────────────────────────────────────┐ │ Step 4: Enhanced Analysis (integrations)│ │ • /impact (change impact) │ │ • /deps (dependencies) │ │ • /security-review (security) │ │ • /multi-ai-review (optional AI) │ └─────────────────────────────────────────┘ ↓ ┌─────────────────────────────────────────┐ │ Step 5: Hook Gate │ │ • policy-gate (permissions + standards│ │ • standards-validator (rules) │ └─────────────────────────────────────────┘ ↓ ┌─────────────────────────────────────────┐ │ Result: Pass / Warning / Fail │ │ • Generate fix guide │ │ • Provide /recover path │ └─────────────────────────────────────────┘ ``` --- ## 2-Stage Review Detail ### Stage 1: Spec Compliance ```yaml Checklist: Requirements match: - Does the changed file's functionality match what is defined in TASKS.md? - Does it match the Mini-PRD / Socrates requirements? Missing checks: - Has the specified error handling been implemented? - Are edge cases handled? YAGNI violations: - Are there unnecessary features not in the spec? - Is there over-engineering? ``` ### Stage 2: Code Quality ```yaml Checklist: Architecture: - SOLID principles - Separation of concerns - Dependency injection Code quality: - Clear naming - Complexity (Cyclomatic, Cognitive) - Code duplication (DRY) - Magic numbers/strings removed Error handling: - All error cases handled - Meaningful error messages - Appropriate logging Testing: - Sufficient coverage - Edge case tests - Tests against real behavior, not just mocks ``` --- ## Severity Classification | Grade | Condition | Action | |-------|-----------|--------| | **Fail** | 1+ Critical issue OR 3+ Important issues | Immediate fix required | | **Warning** | 1–2 Important issues OR many Minor issues | Review before proceeding | | **Pass** | No issues OR Minor only | Proceed to next step | --- ## Ecosystem Integration ### /agile Integration ```yaml /agile (task complete) ↓ /checkpoint auto-invoked ↓ Based on result: - Pass → next task - Warning → user confirmation then proceed - Fail → fix then re-checkpoint ``` ### /team-orchestrate Integration ```yaml /team-orchestrate (task complete) ↓ /checkpoint auto-invoked (post-task gate) ↓ Proceed to next task after Hook gate passes ``` ### PR/Merge Integration ```bash # Auto-invoked from Git Hook pre-commit: /checkpoint --mode=quick # Quick check pre-push: /checkpoint --mode=full # Full check ``` --- ## Security Integration ### /security-review Auto-Invocation ```yaml /security-review invocation conditions: - Changes to auth, payment, or user-related files - Changes to .env or config files - API routing changes Based on result: - Vulnerability found → Fail + fix guide - None → Continue to next stage ``` --- ## AI Multi-Review (Optional) ### /multi-ai-review Integration ```yaml User prompt: "Would you like to include an AI review?" If selected: /multi-ai-review ├── Gemini: code readability, improvement suggestions └── Codex: SOLID, pattern analysis Results integrated into checkpoint report ``` --- ## Output Format ```markdown ## Checkpoint Report ### Overview - **Task**: T1.2 - User authentication API implementation - **Date**: 2026-03-03 15:30 - **Commit**: abc123d ### Change Scope - **Changed files**: 3 - `src/domains/auth/auth.service.ts` (+45, -12) - `src/api/auth.routes.ts` (+23, -5) - `src/middleware/auth.middleware.ts` (+18) ### Stage 1: Spec Compliance ✅ - Requirements match: ✅ - Missing features: ✅ - YAGNI violations: ✅ ### Stage 2: Code Quality ⚠️ - Architecture: ⚠️ Warning - auth.service.ts: Single Responsibility over-scoped - Code quality: ✅ - Testing: ⚠️ Warning - Insufficient edge case coverage ### Integration Analysis - **/impact**: Medium risk (auth-related) - **/deps**: No circular dependencies - **/security-review**: ✅ Passed ### Final Verdict - **Result**: Warning - **Action**: Review then proceed ### Fix Guide 1. Split auth.service.ts into Service + Repository 2. Add edge case tests ``` --- ## Usage Examples ```bash # Basic usage /checkpoint # Specify file scope /checkpoint --files src/auth/*.ts # Include AI multi-review /checkpoint --ai-review # Quick mode (Spec only) /checkpoint --mode=spec # Full mode /checkpoint --mode=full ``` --- **Last Updated**: 2026-03-03 (v1.1.0 - Standalone independent mode complete)