SpecWeave Agents Index

Quick Reference: How to Invoke Agents

All SpecWeave agents follow the naming pattern: {plugin}:{directory}:{name-from-yaml}

Core Planning & Design Agents

Product Manager (PM)

Task({
  subagent_type: "specweave:pm:pm",
  prompt: "Create product requirements for user dashboard feature"
});

Use for: Product strategy, requirements gathering, user story creation, feature prioritization

Architect

Task({
  subagent_type: "specweave:architect:architect",
  prompt: "Design system architecture for user authentication"
});

Use for: System architecture, technical specifications, ADRs, component designs, API contracts

Tech Lead

Task({
  subagent_type: "specweave:tech-lead:tech-lead",
  prompt: "Review code and suggest improvements"
});

Use for: Code review, best practices, technical mentorship, implementation planning

---

Quality & Testing Agents

QA Lead

Task({
  subagent_type: "specweave:qa-lead:qa-lead",
  prompt: "Create test strategy for permission gates feature"
});

Use for: Test plans, test cases, E2E testing with Playwright, test automation, coverage analysis

Test-Aware Planner

Task({
  subagent_type: "specweave:test-aware-planner:test-aware-planner",
  prompt: "Generate tasks with embedded test plans"
});

Use for: Task generation with BDD test plans, coverage targets, AC-test mapping

TDD Orchestrator

Task({
  subagent_type: "specweave:tdd-orchestrator:tdd-orchestrator",
  prompt: "Coordinate TDD workflow for feature implementation"
});

Use for: Red-green-refactor discipline, multi-agent TDD coordination

---

Code Quality & Standards Agents

Code Standards Detective

Task({
  subagent_type: "specweave:code-standards-detective:code-standards-detective",
  prompt: "Analyze codebase and generate coding standards documentation"
});

Use for: Discovering naming conventions, import patterns, detecting anti-patterns

Security

Task({
  subagent_type: "specweave:security:security",
  prompt: "Perform security review of authentication implementation"
});

Use for: Threat modeling, security architecture, OWASP Top 10, vulnerability assessment

---

Documentation & Communication Agents

Docs Writer

Task({
  subagent_type: "specweave:docs-writer:docs-writer",
  prompt: "Create API documentation for REST endpoints"
});

Use for: API docs, user guides, developer guides, README files, architecture documentation

Translator

Task({
  subagent_type: "specweave:translator:AGENT",
  prompt: "Translate documentation to Spanish"
});

Use for: Batch translation projects, multi-file translation coordination

---

Infrastructure & Performance Agents

Infrastructure

Task({
  subagent_type: "specweave:infrastructure:infrastructure",
  prompt: "Generate Terraform configurations for serverless deployment"
});

Use for: Infrastructure-as-Code, AWS Lambda, Azure Functions, GCP Cloud Functions, Supabase

Performance

Task({
  subagent_type: "specweave:performance:performance",
  prompt: "Analyze and optimize database query performance"
});

Use for: Performance optimization, profiling, benchmarking, scalability analysis

---

Quality Assurance Agents

Quality Assessment (Skill-Based)

NOTE: Quality assessment uses the increment-quality-judge-v2 skill (auto-activated), not an agent.

# Use CLI command directly - DO NOT spawn agents
specweave qa 0001 --pre

The skill auto-activates when you discuss quality assessment. See plugins/specweave/skills/increment-quality-judge-v2/SKILL.md.

Reflective Reviewer

Task({
  subagent_type: "specweave:reflective-reviewer:reflective-reviewer",
  prompt: "Perform reflective review of implementation quality"
});

Use for: Post-implementation review, learning from mistakes, improvement suggestions

---

Naming Pattern Explanation

Directory-based agents (most common):

Pattern: {plugin}:{directory}:{name-from-yaml}
Example: specweave:qa-lead:qa-lead

Structure:
plugins/specweave/agents/qa-lead/
  └── AGENT.md (contains: name: qa-lead)

File-based agents (legacy):

Pattern: {plugin}:{filename}
Example: specweave:code-reviewer

Structure:
plugins/specweave/agents/
  └── code-reviewer.md

Finding Agent Types

# List all available agents
ls -la plugins/specweave/agents/

# Check agent's YAML name field
head -5 plugins/specweave/agents/qa-lead/AGENT.md
# Output: name: qa-lead

# Construct full type: specweave:qa-lead:qa-lead

Common Mistakes

// ❌ WRONG: Missing directory/name part
Task({ subagent_type: "specweave:qa-lead", ... });
// Error: Agent type 'specweave:qa-lead' not found

// ✅ CORRECT: Full pattern with directory and name
Task({ subagent_type: "specweave:qa-lead:qa-lead", ... });

See Also

• Agent Details: Each agent's AGENT.md file contains full documentation
• CLAUDE.md: Section 15 - Skills vs Agents: Understanding the Distinction
• Plugin Validation: scripts/validate-plugin-directories.sh