test-spec-gen

Test Specification Generator Skill

Generates production-grade test specifications for any application through multi-agent exploration, research, and specialist generation.

UX Contract

1. User runs /test-spec-gen [optional-filter]
2. Skill enters plan mode, presents approach, gets approval
3. Executes multi-agent pipeline silently
4. Presents test spec document for review
5. Iterates with quick-clarify until satisfied
6. Offers Trello card conversion

Phase 0: Plan Mode Entry (CRITICAL)

HARD GATE: Do NOT proceed with any exploration, research, or generation until plan mode is approved.

Upon invocation:

1. Read existing codebase to understand project type (web/desktop/mobile)
2. Present the multi-agent approach with estimated agent count
3. Ask user approval to proceed
4. Only AFTER approval, begin the exploration phase

Plan Mode Presentation Template

# Test Specification Generator

I will generate a comprehensive test specification for this project using multi-agent orchestration:

## Discovery Phase (5 parallel agents)
- Agent 1: Routes & Pages exploration
- Agent 2: Auth & RBAC mapping
- Agent 3: Data & Backend analysis
- Agent 4: Framework & Config discovery
- Agent 5: Navigation & UX flows

## Research Phase
- /research-before-coding based on discovery findings

## Generation Phase (5 specialist agents)
- Domains determined from exploration + research
- Each generates TC-XXX formatted test cases

## Verification Phase
- Doubt agent review for maintenance burden prevention
- Traceability matrix generation

## Output
- Hermes-style test specification document
- Optional Trello card conversion via /trello-test

Proceed?

EnterPlanMode Trigger

Call EnterPlanMode() immediately after UX Contract section with this prompt:

Generate test specification for [project_name] using multi-agent orchestration:

1. Discovery: 5 parallel explore agents map codebase
2. Research: /research-before-coding for best practices
3. Generation: 5 specialist agents create TC-XXX test cases
4. Verification: Doubt agent review + traceability matrix
5. Output: Test spec doc + optional Trello conversion

Wait for user approval before proceeding to Phase 1.

Phase 0.5: Dependency Pre-flight Check

After plan approval, BEFORE any exploration:

Check Required MCP Servers

# Check if MCP servers are configured
grep -E "perplexity|context7|playwright|trello" ~/.claude/mcp_config.json || echo "MISSING"

Check Test Frameworks

Detect project type and verify:

# Node.js projects
[ -f "package.json" ] && grep -E "(playwright|pytest|vitest|jest)" package.json || echo "NO TEST FRAMEWORK"

# Python projects
[ -f "pyproject.toml" ] && grep -E "(pytest|playwright)" pyproject.toml || echo "NO TEST FRAMEWORK"

Auto-fix with Confirmation

For each missing dependency:

❌ Missing: [dependency]
  Impact: [what breaks without it]
  Auto-fix: [command to install]
  Approve? (Y/n)

If user declines:

• Log the refusal
• Continue only if non-critical
• Refuse to proceed if critical blocker

Critical vs Non-critical

Critical (must refuse without):

• MCP servers for exploration (file read tools)
• Basic test framework detection

Non-critical (can continue with warning):

• Trello API (only needed for card conversion)
• Visual regression tools
• Performance monitoring tools

MCP Discovery Commands

# Check MCP config directly
grep -E "perplexity|context7|playwright|trello" ~/.claude/mcp_config.json 2>/dev/null && echo "MCP servers configured" || echo "Some MCP servers missing"

# Alternative: check ~/.claude.json (Claude Code's actual registry)
grep -E "perplexity|context7|playwright|trello" ~/.claude.json 2>/dev/null | grep -q "mcp" && echo "MCP servers registered" || echo "Some MCP servers missing"

Phase 1: Codebase Discovery (5 Parallel Agents)

Spawn 5 independent explore agents using Task tool. All agents run in parallel with no shared state.

Agent 1: Routes & Pages Explorer

Task(
  subagent_type: "Explore",
  model: "haiku",
  description: "Explore routes and pages",
  prompt: `You are a route and page discovery agent.

OBJECTIVE: Map ALL routes, pages, and UI components in this codebase.

OUTPUT FORMAT (JSON):
{
  "routes": [
    {"path": "/path", "component": "ComponentName", "file": "src/file.tsx", "type": "page|api|static"}
  ],
  "pages": [
    {"name": "Dashboard", "route": "/", "file": "src/pages/Dashboard.tsx", "key_elements": ["stats", "charts"]}
  ],
  "navigation": [
    {"from": "Sidebar", "to": "Dashboard", "label": "Home"}
  ]
}

SEARCH STRATEGY:
1. Find router configuration (App.tsx, router.ts, routes/)
2. Find page components (pages/, views/, screens/)
3. Find navigation components (Nav.tsx, Sidebar.tsx, Header.tsx)
4. Extract route-to-component mappings

Return ONLY the JSON. No explanation.`
)

Agent 2: Auth & RBAC Explorer

Task(
  subagent_type: "Explore",
  model: "haiku",
  description: "Explore auth and RBAC",
  prompt: `You are an authentication and authorization discovery agent.

OBJECTIVE: Map ALL auth mechanisms, roles, permissions, and access controls.

OUTPUT FORMAT (JSON):
{
  "auth_mechanism": "jwt|session|oauth|none",
  "roles": ["admin", "user", "guest"],
  "permissions": ["create:resource", "read:resource"],
  "auth_files": ["src/middleware/auth.ts"],
  "protected_routes": [
    {"route": "/admin", "roles": ["admin"], "guard": "requireAuth"}
  ],
  "login_endpoint": "/api/auth/login"
}

SEARCH STRATEGY:
1. Find auth middleware (auth.ts, middleware/, guards/)
2. Find role definitions (roles.ts, permissions.ts)
3. Find protected route decorators/middleware
4. Find login/logout endpoints

Return ONLY the JSON. No explanation.`
)

Agent 3: Data & Backend Explorer

Task(
  subagent_type: "Explore",
  model: "haiku",
  description: "Explore data and backend",
  prompt: `You are a data and backend discovery agent.

OBJECTIVE: Map ALL data models, API endpoints, and persistence mechanisms.

OUTPUT FORMAT (JSON):
{
  "database": "postgres|mongodb|sqlite|none",
  "orm": "prisma|drizzle|sequelize|none",
  "models": [
    {"name": "User", "fields": ["id", "email", "role"], "file": "models/User.ts"}
  ],
  "api_endpoints": [
    {"method": "GET", "path": "/api/users", "controller": "UserController"}
  ],
  "persistence_files": ["db/schema.prisma"]
}

SEARCH STRATEGY:
1. Find schema definitions (schema.prisma, models/, entities/)
2. Find API routes (api/, routes/, controllers/)
3. Find database config (db.ts, database.ts)
4. Find ORM usage (prisma., drizzle., sequelize.)

Return ONLY the JSON. No explanation.`
)

Agent 4: Framework & Config Explorer

Task(
  subagent_type: "Explore",
  model: "haiku",
  description: "Explore framework and config",
  prompt: `You are a framework and configuration discovery agent.

OBJECTIVE: Map framework, build tooling, CI config, and environment setup.

OUTPUT FORMAT (JSON):
{
  "framework": "react|vue|svelte|next|nuxt|custom",
  "language": "typescript|javascript|python",
  "build_tool": "vite|webpack|rollup|none",
  "test_framework": "playwright|jest|vitest|pytest|none",
  "ci_system": "github-actions|gitlab-ci|none",
  "env_files": [".env", ".env.example"],
  "config_files": ["next.config.js", "vite.config.ts"]
}

SEARCH STRATEGY:
1. Check package.json for dependencies
2. Check for framework config files
3. Check .github/workflows/ for CI
4. Check for .env files

Return ONLY the JSON. No explanation.`
)

Agent 5: Navigation & UX Explorer

Task(
  subagent_type: "Explore",
  model: "haiku",
  description: "Explore navigation and UX",
  prompt: `You are a navigation and user experience discovery agent.

OBJECTIVE: Map navigation flows, state management, and user interactions.

OUTPUT FORMAT (JSON):
{
  "navigation_structure": {
    "main": ["Dashboard", "Settings"],
    "sidebar": ["Items", "Reports"]
  },
  "state_management": "redux|zustand|context|none",
  "k