Native Parallel Agents
Orchestration through Claude Code's built-in Agent Tool
Overview
This skill enables coordinating multiple specialized agents through Claude Code's native agent system. Unlike external scripts, this approach keeps all orchestration within Claude's control.
When to Use Orchestration
✅ Good for:
- Complex tasks requiring multiple expertise domains
- Code analysis from security, performance, and quality perspectives
- Comprehensive reviews (architecture + security + testing)
- Feature implementation needing backend + frontend + database work
❌ Not for:
- Simple, single-domain tasks
- Quick fixes or small changes
- Tasks where one agent suffices
Native Agent Invocation
Single Agent
Use the security-auditor agent to review authentication
Sequential Chain
First, use the explorer-agent to discover project structure.
Then, use the backend-specialist to review API endpoints.
Finally, use the test-engineer to identify test gaps.
With Context Passing
Use the frontend-specialist to analyze React components.
Based on those findings, have the test-engineer generate component tests.
Resume Previous Work
Resume agent [agentId] and continue with additional requirements.
Orchestration Patterns
Pattern 1: Comprehensive Analysis
Agents: explorer-agent → [domain-agents] → synthesis
1. explorer-agent: Map codebase structure
2. security-auditor: Security posture
3. backend-specialist: API quality
4. frontend-specialist: UI/UX patterns
5. test-engineer: Test coverage
6. Synthesize all findings
Pattern 2: Feature Review
Agents: affected-domain-agents → test-engineer
1. Identify affected domains (backend? frontend? both?)
2. Invoke relevant domain agents
3. test-engineer verifies changes
4. Synthesize recommendations
Pattern 3: Security Audit
Agents: security-auditor → penetration-tester → synthesis
1. security-auditor: Configuration and code review
2. penetration-tester: Active vulnerability testing
3. Synthesize with prioritized remediation
Available Agents
| Agent |
Expertise |
Trigger Phrases |
orchestrator |
Coordination |
"comprehensive", "multi-perspective" |
security-auditor |
Security |
"security", "auth", "vulnerabilities" |
penetration-tester |
Security Testing |
"pentest", "red team", "exploit" |
backend-specialist |
Backend |
"API", "server", "Node.js", "Express" |
frontend-specialist |
Frontend |
"React", "UI", "components", "Next.js" |
test-engineer |
Testing |
"tests", "coverage", "TDD" |
devops-engineer |
DevOps |
"deploy", "CI/CD", "infrastructure" |
database-architect |
Database |
"schema", "Prisma", "migrations" |
mobile-developer |
Mobile |
"React Native", "Flutter", "mobile" |
api-designer |
API Design |
"REST", "GraphQL", "OpenAPI" |
debugger |
Debugging |
"bug", "error", "not working" |
explorer-agent |
Discovery |
"explore", "map", "structure" |
documentation-writer |
Documentation |
"write docs", "create README", "generate API docs" |
performance-optimizer |
Performance |
"slow", "optimize", "profiling" |
project-planner |
Planning |
"plan", "roadmap", "milestones" |
seo-specialist |
SEO |
"SEO", "meta tags", "search ranking" |
game-designer |
Game Development |
"game", "Unity", "Godot", "Phaser" |
Claude Code Built-in Agents
These work alongside custom agents:
| Agent |
Model |
Purpose |
| Explore |
Haiku |
Fast read-only codebase search |
| Plan |
Sonnet |
Research during plan mode |
| General-purpose |
Sonnet |
Complex multi-step modifications |
Use Explore for quick searches, custom agents for domain expertise.
Synthesis Protocol
After all agents complete, synthesize:
## Orchestration Synthesis
### Task Summary
[What was accomplished]
### Agent Contributions
| Agent | Finding |
|-------|---------|
| security-auditor | Found X |
| backend-specialist | Identified Y |
### Consolidated Recommendations
1. **Critical**: [Issue from Agent A]
2. **Important**: [Issue from Agent B]
3. **Nice-to-have**: [Enhancement from Agent C]
### Action Items
- [ ] Fix critical security issue
- [ ] Refactor API endpoint
- [ ] Add missing tests
Best Practices
- Available agents - 17 specialized agents can be orchestrated
- Logical order - Discovery → Analysis → Implementation → Testing
- Share context - Pass relevant findings to subsequent agents
- Single synthesis - One unified report, not separate outputs
- Verify changes - Always include test-engineer for code modifications
Key Benefits
- ✅ Single session - All agents share context
- ✅ AI-controlled - Claude orchestrates autonomously
- ✅ Native integration - Works with built-in Explore, Plan agents
- ✅ Resume support - Can continue previous agent work
- ✅ Context passing - Findings flow between agents
1---2name: parallel-agents3description: "Điều phối và chạy song song nhiều mô hình AI agents độc lập cho tác vụ khó."4allowed-tools: Read, Glob, Grep5---67# Native Parallel Agents89> Orchestration through Claude Code's built-in Agent Tool1011## Overview1213This skill enables coordinating multiple specialized agents through Claude Code's native agent system. Unlike external scripts, this approach keeps all orchestration within Claude's control.1415## When to Use Orchestration1617✅ **Good for:**18- Complex tasks requiring multiple expertise domains19- Code analysis from security, performance, and quality perspectives20- Comprehensive reviews (architecture + security + testing)21- Feature implementation needing backend + frontend + database work2223❌ **Not for:**24- Simple, single-domain tasks25- Quick fixes or small changes26- Tasks where one agent suffices2728---2930## Native Agent Invocation3132### Single Agent33```34Use the security-auditor agent to review authentication35```3637### Sequential Chain38```39First, use the explorer-agent to discover project structure.40Then, use the backend-specialist to review API endpoints.41Finally, use the test-engineer to identify test gaps.42```4344### With Context Passing45```46Use the frontend-specialist to analyze React components.47Based on those findings, have the test-engineer generate component tests.48```4950### Resume Previous Work51```52Resume agent [agentId] and continue with additional requirements.53```5455---5657## Orchestration Patterns5859### Pattern 1: Comprehensive Analysis60```61Agents: explorer-agent → [domain-agents] → synthesis62631. explorer-agent: Map codebase structure642. security-auditor: Security posture653. backend-specialist: API quality664. frontend-specialist: UI/UX patterns675. test-engineer: Test coverage686. Synthesize all findings69```7071### Pattern 2: Feature Review72```73Agents: affected-domain-agents → test-engineer74751. Identify affected domains (backend? frontend? both?)762. Invoke relevant domain agents773. test-engineer verifies changes784. Synthesize recommendations79```8081### Pattern 3: Security Audit82```83Agents: security-auditor → penetration-tester → synthesis84851. security-auditor: Configuration and code review862. penetration-tester: Active vulnerability testing873. Synthesize with prioritized remediation88```8990---9192## Available Agents9394| Agent | Expertise | Trigger Phrases |95|-------|-----------|-----------------|96| `orchestrator` | Coordination | "comprehensive", "multi-perspective" |97| `security-auditor` | Security | "security", "auth", "vulnerabilities" |98| `penetration-tester` | Security Testing | "pentest", "red team", "exploit" |99| `backend-specialist` | Backend | "API", "server", "Node.js", "Express" |100| `frontend-specialist` | Frontend | "React", "UI", "components", "Next.js" |101| `test-engineer` | Testing | "tests", "coverage", "TDD" |102| `devops-engineer` | DevOps | "deploy", "CI/CD", "infrastructure" |103| `database-architect` | Database | "schema", "Prisma", "migrations" |104| `mobile-developer` | Mobile | "React Native", "Flutter", "mobile" |105| `api-designer` | API Design | "REST", "GraphQL", "OpenAPI" |106| `debugger` | Debugging | "bug", "error", "not working" |107| `explorer-agent` | Discovery | "explore", "map", "structure" |108| `documentation-writer` | Documentation | "write docs", "create README", "generate API docs" |109| `performance-optimizer` | Performance | "slow", "optimize", "profiling" |110| `project-planner` | Planning | "plan", "roadmap", "milestones" |111| `seo-specialist` | SEO | "SEO", "meta tags", "search ranking" |112| `game-designer` | Game Development | "game", "Unity", "Godot", "Phaser" |113114---115116## Claude Code Built-in Agents117118These work alongside custom agents:119120| Agent | Model | Purpose |121|-------|-------|---------|122| **Explore** | Haiku | Fast read-only codebase search |123| **Plan** | Sonnet | Research during plan mode |124| **General-purpose** | Sonnet | Complex multi-step modifications |125126Use **Explore** for quick searches, **custom agents** for domain expertise.127128---129130## Synthesis Protocol131132After all agents complete, synthesize:133134```markdown135## Orchestration Synthesis136137### Task Summary138[What was accomplished]139140### Agent Contributions141| Agent | Finding |142|-------|---------|143| security-auditor | Found X |144| backend-specialist | Identified Y |145146### Consolidated Recommendations1471. **Critical**: [Issue from Agent A]1482. **Important**: [Issue from Agent B]1493. **Nice-to-have**: [Enhancement from Agent C]150151### Action Items152- [ ] Fix critical security issue153- [ ] Refactor API endpoint154- [ ] Add missing tests155```156157---158159## Best Practices1601611. **Available agents** - 17 specialized agents can be orchestrated1622. **Logical order** - Discovery → Analysis → Implementation → Testing1633. **Share context** - Pass relevant findings to subsequent agents1644. **Single synthesis** - One unified report, not separate outputs1655. **Verify changes** - Always include test-engineer for code modifications166167---168169## Key Benefits170171- ✅ **Single session** - All agents share context172- ✅ **AI-controlled** - Claude orchestrates autonomously173- ✅ **Native integration** - Works with built-in Explore, Plan agents174- ✅ **Resume support** - Can continue previous agent work175- ✅ **Context passing** - Findings flow between agents
tuyenht