verification-before-completion skill

---
name: verifying-before-completion
description: AI agent validates work through evidence-based verification with comprehensive checklists, automated validation, and proof-of-work documentation. Use when completing tasks, preparing for review, or ensuring quality.
---

# Verifying Before Completion

## Quick Start

1. **Functional** - All acceptance criteria met, happy path works, edge cases handled
2. **Technical** - Tests pass, coverage met, lint clean, build succeeds
3. **Security** - No secrets exposed, input sanitized, auth verified
4. **Performance** - Response time acceptable, no N+1, no memory leaks
5. **Documentation** - Code comments, API docs, README updated
6. **Evidence** - Collect screenshots, test output, metrics for proof

## Features

| Feature | Description | Guide |
|---------|-------------|-------|
| Universal Checklist | Comprehensive validation | Functional, technical, security, perf, docs |
| Evidence Collection | Proof of verification | Test output, screenshots, metrics |
| Automated Pipeline | CI/CD validation | Type check, lint, test, build, audit |
| Type-Specific Checks | Context-aware validation | UI, API, migration, bug fix, optimization |
| Pre-Commit Hooks | Catch issues early | lint-staged, related tests |
| Sign-Off Template | Formal completion record | Verified by, date, evidence links |

## Common Patterns

```
# Universal Verification Checklist
FUNCTIONAL:
[ ] All acceptance criteria met
[ ] Happy path works correctly
[ ] Edge cases handled
[ ] Error handling works
[ ] Input validation works

TECHNICAL:
[ ] Tests pass (unit, integration, E2E)
[ ] Coverage meets threshold (80%+)
[ ] No TypeScript/lint errors
[ ] Build succeeds
[ ] No security vulnerabilities

DOCUMENTATION:
[ ] Complex logic commented
[ ] API documentation updated
[ ] README updated if needed
```

```bash
# Automated Verification Script
npm run lint         # Lint check
npm run typecheck    # Type check
npm test             # Unit tests
npm run build        # Build verification
npm audit            # Security audit

# Full verification
npm run verify       # Runs all checks
```

```
# Evidence Documentation Template
## Test Evidence
$ npm test
Test Suites: 15 passed
Tests: 87 passed
Coverage: 85.2%

## Manual Verification
| Step | Expected | Actual | Status |
|------|----------|--------|--------|
| Login | Form appears | Works | Pass |
| Submit | Success msg | Shows | Pass |

## Performance Metrics
| Metric | Baseline | Current | Status |
|--------|----------|---------|--------|
| Bundle | 245 KB | 248 KB | Pass |
| API | 150ms | 145ms | Pass |

## Sign-Off
[x] All tests pass
[x] Manual testing complete
[x] Ready for review
Verified by: [Name] on [Date]
```

```
# Type-Specific Checklists
UI COMPONENT:
[ ] Renders in all browsers
[ ] Responsive mobile/tablet/desktop
[ ] Loading/error/empty states
[ ] Keyboard accessible
[ ] Matches design specs

API ENDPOINT:
[ ] Correct status codes
[ ] Response matches schema
[ ] Auth/authz enforced
[ ] Input validated
[ ] Documented in OpenAPI

BUG FIX:
[ ] Root cause identified
[ ] Fix addresses root cause
[ ] Regression test added
[ ] No new bugs introduced
```

## Best Practices

| Do | Avoid |
|----|-------|
| Run all tests before marking complete | Skipping verification steps |
| Document evidence of verification | Marking complete without evidence |
| Test both happy and unhappy paths | Assuming tests are enough |
| Verify on multiple browsers/devices | Ignoring edge cases |
| Check performance impact | Skipping security checks |
| Use automated verification | Merging without green CI |
| Include screenshots for UI | Forgetting mobile testing |
| Self-review before requesting | Rushing verification |

## Related Skills

- `finishing-development-branches` - Complete branch preparation
- `executing-plans` - Quality gates during execution
- `debugging-systematically` - Verify fixes work
- `developing-test-driven` - Build verification into process