Skip to content
All Skills

Validate Spec

Validate spec completeness and implementation compliance. Use before approval (--pre) or before completion (--post).

Software Engineering|v1|Updated 7/14/2026|GitHub source
MCP get_skill({ skillId: "validate-spec-ab07e4c3" })

Use this skill with your agent

Create a free account and connect via MCP

Get Started Free
# /validate-spec

Validate spec quality (pre-implementation) or implementation compliance (post-implementation).

## Usage

```bash
/validate-spec SPEC-001              # Auto-detect mode based on status
/validate-spec SPEC-001 --pre        # Pre-implementation validation
/validate-spec SPEC-001 --post       # Post-implementation compliance
```

## Mode Detection

| Spec Status | Default Mode | Purpose |
|-------------|--------------|---------|
| draft, review | --pre | Validate before approval |
| approved | --pre | Final check before implementation |
| in_progress, complete | --post | Verify implementation matches |

## Pre-Implementation Validation (--pre)

### 1. Structure Completeness
```
Required sections:
✓ Summary (present)
✓ Problem Statement (present)
✓ Acceptance Scenarios (3 found)
✓ Agent Context (present)
✓ Out of Scope (present)
✗ Dependencies (MISSING)
```

### 2. Acceptance Scenario Quality
```
Scenario 1: "User signs up"
✓ Has Given/When/Then clauses
✓ Specific, not vague

Scenario 2: "Invalid credentials"
⚠ Then clause is vague: "shows error"
  Suggest: "shows 'Invalid email or password' message"
```

Scenarios must be:
- **Specific**: Not "shows error" but "shows error X"
- **Testable**: Can become automated test
- **Complete**: Given/When/Then present

### 3. Agent Context Validation
```
Files referenced:
✓ src/lib/auth.ts exists
✗ src/middleware/session.ts NOT FOUND
```

### 4. Success Metrics Measurability
```
Metric 1: "Page load < 2s"
✓ Measurable, has threshold

Metric 2: "Good user experience"
✗ NOT MEASURABLE
  Suggest: "User satisfaction > 4/5"
```

### 5. Conciseness Check
```
Word count: 847 words
Guideline: < 500 words
⚠ Consider splitting into multiple specs
```

### 6. Open Questions Check
```
2 open questions found:
1. "Which OAuth providers?"
2. "Session timeout?"

⚠ Resolve before 'approved' status
```

### Pre-Validation Summary
```
Ready for approval: NO

Fix:
- Add missing Dependencies section
- Refine vague scenario
- Make metrics measurable
- Resolve open questions
```

## Post-Implementation Validation (--post)

### 1. Acceptance Scenario Coverage
```
Scenario: "User signs up"
✓ Found: auth/signup.test.ts:15

Scenario: "Session expires"
✗ NO TEST FOUND

Coverage: 2/3 scenarios (67%)
```

### 2. Success Metrics Verification
```
Metric: "Page load < 2s"
  Result: 1.4s average
  ✓ PASS

Metric: "95% coverage"
  Result: 97.2%
  ✓ PASS
```

### 3. Out-of-Scope Violations
```
Out of scope: "Social login", "2FA"

Scanning...
✓ No oauth/social code found
⚠ Found "password strength" - potential violation
```

### 4. Agent Constraints Compliance
```
Constraint: "No new dependencies"
✓ package.json unchanged

Constraint: "Files outside scope"
⚠ src/utils/helpers.ts modified
```

### 5. Definition of Done
```
- [x] Acceptance scenarios tested
- [ ] No console errors ← incomplete
```

### Post-Validation Summary
```
Implementation compliance: PARTIAL

Issues to resolve:
1. Add test for "Session expires"
2. Review password complexity code
3. Fix console warnings
```

## Interactive Fixes

When issues found:
```
AI: Let's refine Scenario 2...
    What should the error message say?

User: "Invalid email or password"

AI: Updated. Now the metrics...
    Remove "Good UX" or replace with measurable?

User: Remove it

AI: Done. Re-validating...

    ## Validation: PASS
    Ready to move to 'review' status?
```

## Integration

```
/spec → /validate-spec --pre → [approval] → /plan → /implement → /validate-spec --post → /complete
```
#personal-productivity#daily-routine#weekly-review#life-planning#personal-knowledge#testing

Related Skills

More skills in Software Engineering

Accessibility Standards

Comprehensive web accessibility standards based on WCAG 2.2 AA, with 38+ anti-patterns, legal enforcement context (EAA, ADA Title II), WAI-ARIA patterns, and framework-specific fixes for modern web frameworks and libraries.

#github-copilot#accessibilityMIT

Accord

Authoring unified specification packages across Business/Development/Design teams via staged elaboration (L0 Vision → L1 Requirements → L2 Team Detail → L3 Acceptance Criteria). No code. Use when authoring cross-team specs, building L0-L3 packages, or aligning Biz/Dev/Design on a single source of truth.

#broad-capability#developmentMIT

Acquire Codebase Knowledge

Use this skill when the user explicitly asks to map, document, or onboard into an existing codebase. Trigger for prompts like "map this codebase", "document this architecture", "onboard me to this repo", or "create codebase docs". Do not trigger for routine feature implementation, bug fixes, or narrow code edits unless the user asks for repository-level discovery.

#github-copilot#documentationMIT

Acreadiness Assess

Run the AgentRC readiness assessment on the current repository and produce a static HTML dashboard at reports/index.html. Wraps `npx github:microsoft/agentrc readiness` and hands off rendering to the @ai-readiness-reporter custom agent. Supports policies (--policy) for org-specific scoring. Use when asked to assess, audit, or score the AI readiness of a repo.

#github-copilot#planningMIT

Acreadiness Generate Instructions

Generate tailored AI agent instruction files via AgentRC instructions command. Produces .github/copilot-instructions.md (default, recommended for Copilot in VS Code) plus optional per-area .instructions.md files with applyTo globs for monorepos. Use after running /acreadiness-assess to close gaps in the AI Tooling pillar.

#github-copilot#skillMIT

Acreadiness Policy

Help the user pick, write, or apply an AgentRC policy. Policies customise readiness scoring by disabling irrelevant checks, overriding impact/level, setting pass-rate thresholds, or chaining org baselines with team overrides. Use when the user asks about strict mode, AI-only scoring, custom weights, CI gating, or wants org-wide standardisation.

#github-copilot#planningMIT