Read tasks.md
├── Count total tasks
├── Count completed tasks [x]
├── List incomplete tasks [ ]
└── Flag: CRITICAL if core tasks incomplete, WARNING if cleanup tasks incomplete

Build command detection:
  openspec/config.yaml → rules.verify.build_command
  package.json → "scripts.build"
  Makefile → "build" target
  Otherwise → skip build check

Test command detection:
  openspec/config.yaml → rules.verify.test_command
  openspec/config.yaml → rules.apply.test_command
  package.json → "scripts.test"
  pytest.ini / pyproject.toml → pytest
  Makefile → "test" target
  Otherwise → skip test run

Coverage threshold:
  openspec/config.yaml → rules.verify.coverage_threshold (default: 0 = disabled)

AUTOMATED CHECKS (attempt all — report result for each):
├── Linter: e.g., npm run lint, ruff check, biome check
├── Type checker: e.g., tsc --noEmit, mypy, dotnet build
├── Build: {detected build_command}
├── Tests: {detected test_command} — run ALL tests, not just new ones
└── Coverage: if threshold > 0, check coverage report against threshold

FOR EACH REQUIREMENT in specs/:
├── Search codebase for implementation evidence
├── Search test files for tests covering each SCENARIO
├── Cross-reference with test results from Step 2b
└── Classify as:
    ✅ COMPLIANT     — implemented + tested + test passed
    ⚠️ PARTIAL       — implemented but test missing or test failed
    ❌ NOT COMPLIANT — not implemented
    ⏭ SKIPPED       — intentionally deferred (must be noted in proposal/tasks)

FOR EACH DECISION in design.md:
├── Was the chosen approach actually used?
├── Were rejected alternatives accidentally implemented?
├── Do file changes match the "File Changes" table?
└── Flag: WARNING if deviation found (may be valid improvement)

SECURITY AUDIT:
├── Are there hardcoded secrets, keys, or passwords?
├── Is user input validated/sanitized before use?
├── Are new dependencies from trusted sources?
├── Are new API endpoints properly authenticated/authorized?
└── Flag: CRITICAL for security issues, WARNING for best-practice gaps

REGRESSION CHECK:
├── Do existing tests still pass? (from Step 2b full test run)
├── Are there breaking changes to public APIs or interfaces?
├── Do file deletions leave dangling imports/references?
└── Flag: CRITICAL for regressions, WARNING for potential issues

## Verification Report

**Change**: {change-name}
**Persistence**: {engram (ID: #{id}) | openspec (path) | none (inline)}

### Completeness
| Metric | Value |
|--------|-------|
| Tasks total | {N} |
| Tasks complete | {N} |
| Tasks incomplete | {N} |

{List incomplete tasks if any}

### Automated Checks
| Check | Result | Details |
|-------|--------|--------|
| Linter | ✅ Pass / ❌ Fail / ⏭ Skipped | {details if failed or reason if skipped} |
| Type Check | ✅ Pass / ❌ Fail / ⏭ Skipped | {details if failed} |
| Build | ✅ Pass / ❌ Fail / ⏭ Skipped | {error details or reason if skipped} |
| Tests | ✅ X passed / ❌ Y failed / ⏭ Skipped | {failing test names or reason if skipped} |
| Coverage | ✅ X% (≥ threshold) / ❌ X% (< threshold) / ⏭ Disabled | |

### Spec Compliance Matrix
| Requirement | Implemented | Tests Exist | Tests Pass | Status |
|------------|-------------|-------------|------------|--------|
| REQ-XXX-001: {name} | ✅ / ❌ | ✅ / ❌ | ✅ / ❌ / ⏭ | ✅ COMPLIANT / ⚠️ PARTIAL / ❌ NOT COMPLIANT |
| REQ-XXX-002: {name} | ✅ / ❌ | ✅ / ❌ | ✅ / ❌ / ⏭ | ... |
| NFR-001: {name} | ✅ / ❌ | ✅ / ❌ | ✅ / ❌ / ⏭ | ... |

**Scenario Coverage:**
| Scenario | Spec Req | Test Exists | Test Passed | Status |
|----------|----------|-------------|-------------|--------|
| {scenario name} | REQ-XXX-001 | ✅ | ✅ | ✅ COMPLIANT |
| {scenario name} | REQ-XXX-002 | ✅ | ❌ | ⚠️ PARTIAL |
| {scenario name} | REQ-XXX-003 | ❌ | ⏭ | ❌ NOT COMPLIANT |

### Coherence (Design)
| Decision | Followed? | Notes |
|----------|-----------|-------|
| {Decision name} | ✅ Yes | |
| {Decision name} | ⚠️ Deviated | {how and why} |

### Security & Regression
| Check | Status | Notes |
|-------|--------|-------|
| Hardcoded secrets | ✅ None / ❌ Found | {details} |
| Input validation | ✅ OK / ⚠️ Gaps | {details} |
| Auth on new endpoints | ✅ OK / ⚠️ Missing | {details} |
| Breaking changes | ✅ None / ⚠️ Found | {details} |
| Dangling references | ✅ None / ⚠️ Found | {details} |

### Issues Found

**CRITICAL** (must fix before archive):
{List or "None"}

**WARNING** (should fix):
{List or "None"}

**SUGGESTION** (nice to have):
{List or "None"}

### Verdict
{PASS / PASS WITH WARNINGS / FAIL}

{One-line summary of overall status}