A curated list of agent files and resources developed with Claude Code's Sub-Agent functionality, designed to enhance development workflows with domain-specific expertise.
🧠 Answer: Sub Agents are an amazing new feature offered by Claude Code and the folks at Anthropic. Per their website:
Custom sub agents in Claude Code are specialized AI assistants that can be invoked to handle specific types of tasks. They enable more efficient problem-solving by providing task-specific configurations with customized system prompts, tools and a separate context window.
This repository contains specialized subagents that extend Claude Code's capabilities. Each subagent is an expert in a specific domain, automatically invoked based on context or explicitly called when needed. All agents are configured with specific Claude models based on task complexity for optimal performance and cost-effectiveness.
📋 63 specialized subagents organized across 16 categories
See the complete directory: AGENTS.md
- 🎯 Orchestration (8) - Multi-agent coordination, workflows, task distribution
- 🔬 Research (4) - Market analysis, competitive intelligence, data research
- ⚡ Development (6) - Core development, debugging, refactoring, tooling
- 🏗️ Architecture (2) - System design, cloud architecture, code review
- 🤖 Data-AI-ML (9) - Machine learning, data engineering, AI systems
- ☁️ DevOps-Infrastructure (2) - Operations, networking, infrastructure automation
- 💼 Business (4) - Customer success, legal, support, trend analysis
- 🎨 Design-UX (5) - User experience, visual design, design systems
- 📚 Documentation (3) - Technical writing, tutorials, API documentation
- 🔒 Security (1) - Security auditing, vulnerability assessment
- ⚡ Performance (1) - Performance optimization, load testing
- 📱 Mobile-Frontend (3) - Frontend development, mobile apps, JavaScript
- 🔧 Backend (9) - APIs, databases, microservices, server-side
- 📋 Management (3) - Project management, product strategy, business analysis
- 📢 Sales-Marketing (3) - Content marketing, sales automation, technical sales
- 🚀 MarTech (7) - Marketing technology, automation, personalization
These subagents are automatically available when placed in ~/.claude/agents/ directory.
cd ~/.claude
git clone https://github.com/melanysft/claude-code-subagents.git agentsClaude Code will automatically delegate to the appropriate subagent based on the task context and the subagent's description.
Mention the subagent by name in your request:
"Use the code-reviewer to check my recent changes"
"Have the security-auditor scan for vulnerabilities"
"Get the performance-engineer to optimize this bottleneck"
# Code quality and review
"Use code-reviewer to analyze this component for best practices"
"Have code-reviewer scrutinize these configuration changes"
"Have security-auditor check for OWASP compliance issues"
# Development tasks
"Get backend-architect to design a user authentication API"
"Use frontend-developer to create a responsive dashboard layout"
# Infrastructure and operations
"Have devops-troubleshooter analyze these production logs"
"Use cloud-architect to design a scalable AWS architecture"
"Get network-engineer to debug SSL certificate issues"
"Use database-admin to set up backup and replication"
# Data and AI
"Get data-scientist to analyze this customer behavior dataset"
"Use ai-engineer to build a RAG system for document search"
"Have mlops-engineer set up MLflow experiment tracking"
# Business and marketing
"Have business-analyst create investor deck with growth metrics"
"Use content-marketer to write SEO-optimized blog post"
"Get sales-automator to create cold email sequence"
"Have customer-support draft FAQ documentation"These subagents work together seamlessly, and for more complex orchestrations, you can use the Claude Code Commands collection which provides 52 pre-built slash commands that leverage these subagents in sophisticated workflows.
# Feature development workflow
"Implement user authentication feature"
# Automatically uses: backend-architect → frontend-developer → test-automator → security-auditor
# Performance optimization workflow
"Optimize the checkout process performance"
# Automatically uses: performance-engineer → database-optimizer → frontend-developer
# Production incident workflow
"Debug high memory usage in production"
# Automatically uses: incident-responder → devops-troubleshooter → error-detective → performance-engineer
# Network connectivity workflow
"Fix intermittent API timeouts"
# Automatically uses: network-engineer → devops-troubleshooter → performance-engineer
# Database maintenance workflow
"Set up disaster recovery for production database"
# Automatically uses: database-admin → database-optimizer → incident-responder
# ML pipeline workflow
"Build end-to-end ML pipeline with monitoring"
# Automatically uses: mlops-engineer → ml-engineer → data-engineer → performance-engineer
# Product launch workflow
"Launch new feature with marketing campaign"
# Automatically uses: business-analyst → content-marketer → sales-automator → customer-supportFor more sophisticated multi-subagent orchestration, use the companion Commands repository:
# Complex feature development (8+ subagents)
/full-stack-feature Build user dashboard with real-time analytics
# Production incident response (5+ subagents)
/incident-response Database connection pool exhausted
# ML infrastructure setup (6+ subagents)
/ml-pipeline Create recommendation engine with A/B testing
# Security-focused implementation (7+ subagents)
/security-hardening Implement OAuth2 with zero-trust architectureEach subagent follows this structure:
---
name: subagent-name
description: When this subagent should be invoked
model: haiku # Optional - specify which model to use (haiku/sonnet/opus)
tools: tool1, tool2 # Optional - defaults to all tools
---
System prompt defining the subagent's role and capabilitiesAs of Claude Code v1.0.64, subagents can specify which Claude model they should use. This allows for cost-effective task delegation based on complexity:
- Low Complexity (Haiku): Simple tasks like basic data analysis, documentation generation, and standard responses
- Medium Complexity (Sonnet): Development tasks, code review, testing, and standard engineering work
- High Complexity (Opus): Critical tasks like security auditing, architecture review, incident response, and AI/ML engineering
Available models (using simplified naming as of Claude Code v1.0.64):
haiku- Fast and cost-effective for simple taskssonnet- Balanced performance for most development workopus- Most capable for complex analysis and critical tasks
If no model is specified, the subagent will use the system's default model.
Claude Code automatically coordinates agents using these common patterns:
User Request → Agent A → Agent B → Agent C → Result
Example: "Build a new API feature"
backend-architect → frontend-developer → test-automator → security-auditor
User Request → Agent A + Agent B (simultaneously) → Merge Results
Example: "Optimize application performance"
performance-engineer + database-optimizer → Combined recommendations
User Request → Analysis → Route to appropriate specialist
Example: "Fix this bug"
debugger (analyzes) → Routes to: backend-architect OR frontend-developer OR devops-troubleshooter
Primary Agent → Review Agent → Final Result
Example: "Implement payment processing"
payment-integration → security-auditor → Validated implementation
- Let Claude Code delegate automatically - The main agent analyzes context and selects optimal agents
- Be specific about requirements - Include constraints, tech stack, and quality requirements
- Trust agent expertise - Each agent is optimized for their domain
- Start with high-level requests - Let agents coordinate complex multi-step tasks
- Provide context between agents - Ensure agents have necessary background information
- Review integration points - Check how different agents' outputs work together
- Use explicit invocation for specific needs - When you want a particular expert's perspective
- Combine multiple agents strategically - Different specialists can validate each other's work
- Request specific review patterns - "Have security-auditor review backend-architect's API design"
- Monitor agent effectiveness - Learn which agents work best for your use cases
- Iterate on complex tasks - Use agent feedback to refine requirements
- Leverage agent strengths - Match task complexity to agent capabilities
To add a new subagent:
- Create a new
.mdfile following the format above - Use lowercase, hyphen-separated names
- Write clear descriptions for when the subagent should be used
- Include specific instructions in the system prompt
Agent not being invoked automatically:
- Ensure your request clearly indicates the domain (e.g., "performance issue" → performance-engineer)
- Be specific about the task type (e.g., "review code" → code-reviewer)
Unexpected agent selection:
- Provide more context about your tech stack and requirements
- Use explicit invocation if you need a specific agent
Multiple agents producing conflicting advice:
- This is normal - different specialists may have different priorities
- Ask for clarification: "Reconcile the recommendations from security-auditor and performance-engineer"
Agent seems to lack context:
- Provide background information in your request
- Reference previous conversations or established patterns
If agents aren't working as expected:
- Check agent descriptions in their individual files
- Try more specific language in your requests
- Use explicit invocation to test specific agents
- Provide more context about your project and goals
This project is licensed under the MIT License - see the LICENSE file for details.