agents/README.md

Claude Code Subagents Collection

A comprehensive collection of 78 specialized AI subagents for Claude Code, providing domain-specific expertise across software development, infrastructure, and business operations.

Overview

This repository provides production-ready subagents that extend Claude Code's capabilities with specialized knowledge. Each subagent incorporates:

• Current industry best practices and standards (2024/2025)
• Production-ready patterns and enterprise architectures
• Deep domain expertise with 8-12 capability areas per agent
• Modern technology stacks and frameworks
• Optimized model selection based on task complexity

Agent Categories

Architecture & System Design

Core Architecture

• backend-architect - RESTful API design, microservice boundaries, database schemas
• frontend-developer - React components, responsive layouts, client-side state management
• graphql-architect - GraphQL schemas, resolvers, federation architecture
• architect-reviewer - Architectural consistency analysis and pattern validation
• cloud-architect - AWS/Azure/GCP infrastructure design and cost optimization
• hybrid-cloud-architect - Multi-cloud strategies across cloud and on-premises environments
• kubernetes-architect - Cloud-native infrastructure with Kubernetes and GitOps

UI/UX & Mobile

• ui-ux-designer - Interface design, wireframes, design systems
• ui-visual-validator - Visual regression testing and UI verification
• mobile-developer - React Native and Flutter application development
• ios-developer - Native iOS development with Swift/SwiftUI
• flutter-expert - Advanced Flutter development with state management

Programming Languages

Systems & Low-Level

• c-pro - System programming with memory management and OS interfaces
• cpp-pro - Modern C++ with RAII, smart pointers, STL algorithms
• rust-pro - Memory-safe systems programming with ownership patterns
• golang-pro - Concurrent programming with goroutines and channels

Web & Application

• javascript-pro - Modern JavaScript with ES6+, async patterns, Node.js
• typescript-pro - Advanced TypeScript with type systems and generics
• python-pro - Python development with advanced features and optimization
• ruby-pro - Ruby with metaprogramming, Rails patterns, gem development
• php-pro - Modern PHP with frameworks and performance optimization

Enterprise & JVM

• java-pro - Modern Java with streams, concurrency, JVM optimization
• scala-pro - Enterprise Scala with functional programming and distributed systems
• csharp-pro - C# development with .NET frameworks and patterns

Specialized Platforms

• elixir-pro - Elixir with OTP patterns and Phoenix frameworks
• unity-developer - Unity game development and optimization
• minecraft-bukkit-pro - Minecraft server plugin development
• sql-pro - Complex SQL queries and database optimization

Infrastructure & Operations

DevOps & Deployment

• devops-troubleshooter - Production debugging, log analysis, deployment troubleshooting
• deployment-engineer - CI/CD pipelines, containerization, cloud deployments
• terraform-specialist - Infrastructure as Code with Terraform modules and state management
• dx-optimizer - Developer experience optimization and tooling improvements

Database Management

• database-optimizer - Query optimization, index design, migration strategies
• database-admin - Database operations, backup, replication, monitoring

Incident Response & Network

• incident-responder - Production incident management and resolution
• network-engineer - Network debugging, load balancing, traffic analysis

Quality Assurance & Security

Code Quality & Review

• code-reviewer - Code review with security focus and production reliability
• security-auditor - Vulnerability assessment and OWASP compliance
• architect-reviewer - Architectural consistency and pattern validation

Testing & Debugging

• test-automator - Comprehensive test suite creation (unit, integration, e2e)
• tdd-orchestrator - Test-Driven Development methodology guidance
• debugger - Error resolution and test failure analysis
• error-detective - Log analysis and error pattern recognition

Performance & Research

• performance-engineer - Application profiling and optimization
• search-specialist - Advanced web research and information synthesis

Data & AI

Data Engineering & Analytics

• data-scientist - Data analysis, SQL queries, BigQuery operations
• data-engineer - ETL pipelines, data warehouses, streaming architectures

Machine Learning & AI

• ai-engineer - LLM applications, RAG systems, prompt pipelines
• ml-engineer - ML pipelines, model serving, feature engineering
• mlops-engineer - ML infrastructure, experiment tracking, model registries
• prompt-engineer - LLM prompt optimization and engineering

Documentation & Technical Writing

• docs-architect - Comprehensive technical documentation generation
• api-documenter - OpenAPI/Swagger specifications and developer docs
• reference-builder - Technical references and API documentation
• tutorial-engineer - Step-by-step tutorials and educational content
• mermaid-expert - Diagram creation (flowcharts, sequences, ERDs)

Business & Operations

Business Analysis & Finance

• business-analyst - Metrics analysis, reporting, KPI tracking
• quant-analyst - Financial modeling, trading strategies, market analysis
• risk-manager - Portfolio risk monitoring and management

Marketing & Sales

• content-marketer - Blog posts, social media, email campaigns
• sales-automator - Cold emails, follow-ups, proposal generation

Support & Legal

• customer-support - Support tickets, FAQ responses, customer communication
• hr-pro - HR operations, policies, employee relations
• legal-advisor - Privacy policies, terms of service, legal documentation

Specialized Domains

• payment-integration - Payment processor integration (Stripe, PayPal)
• legacy-modernizer - Legacy code refactoring and modernization
• context-manager - Multi-agent context management

SEO & Content Optimization

• seo-content-auditor - Content quality analysis, E-E-A-T signals assessment
• seo-meta-optimizer - Meta title and description optimization
• seo-keyword-strategist - Keyword analysis and semantic variations
• seo-structure-architect - Content structure and schema markup
• seo-snippet-hunter - Featured snippet formatting
• seo-content-refresher - Content freshness analysis
• seo-cannibalization-detector - Keyword overlap detection
• seo-authority-builder - E-E-A-T signal analysis
• seo-content-writer - SEO-optimized content creation
• seo-content-planner - Content planning and topic clusters

Model Configuration

Agents are assigned to specific Claude models based on task complexity and computational requirements. The system uses three model tiers:

Haiku Model (Fast Response) - 11 agents

Optimized for quick, focused tasks with minimal computational overhead:

• context-manager, reference-builder, sales-automator, search-specialist
• SEO agents: seo-meta-optimizer, seo-keyword-strategist, seo-structure-architect, seo-snippet-hunter, seo-content-refresher, seo-cannibalization-detector, seo-content-planner

Sonnet Model (Balanced Performance) - 46 agents

Standard model for development and specialized tasks:

• Language specialists (18): All programming language agents
• Frontend/UI agents (5): frontend-developer, ui-ux-designer, ui-visual-validator, mobile-developer, ios-developer
• Infrastructure agents (14): DevOps, database, network, and deployment specialists
• Quality/Testing agents (4): test-automator, tdd-orchestrator, debugger, error-detective
• Data/ML agents (5): Basic ML and data engineering tasks

Opus Model (Maximum Capability) - 21 agents

Reserved for complex reasoning and critical analysis:

• Architecture & Design (7): architect-reviewer, backend-architect, cloud-architect, hybrid-cloud-architect, kubernetes-architect, graphql-architect, terraform-specialist
• Critical Analysis (5): code-reviewer, security-auditor, performance-engineer, incident-responder, database-optimizer
• AI/ML Complex (5): ai-engineer, ml-engineer, mlops-engineer, data-scientist, prompt-engineer
• Business Critical (4): docs-architect, hr-pro, legal-advisor, quant-analyst

Installation

Clone the repository to the Claude agents directory:

cd ~/.claude
git clone https://github.com/wshobson/agents.git

The subagents will be automatically available to Claude Code once placed in the ~/.claude/agents/ directory.

Usage

Automatic Delegation

Claude Code automatically selects the appropriate subagent based on task context and requirements. The system analyzes your request and delegates to the most suitable specialist.

Explicit Invocation

Specify a subagent by name to use a particular specialist:

"Use code-reviewer to analyze the recent changes"
"Have security-auditor scan for vulnerabilities"
"Get performance-engineer to optimize this bottleneck"

Usage Examples

Code Quality & Security

code-reviewer: Analyze component for best practices
security-auditor: Check for OWASP compliance
tdd-orchestrator: Implement feature with test-first approach
performance-engineer: Profile and optimize bottlenecks

Development & Architecture

backend-architect: Design authentication API
frontend-developer: Create responsive dashboard
graphql-architect: Design federated GraphQL schema
mobile-developer: Build cross-platform mobile app

Infrastructure & Operations

devops-troubleshooter: Analyze production logs
cloud-architect: Design scalable AWS architecture
network-engineer: Debug SSL certificate issues
database-admin: Configure backup and replication
terraform-specialist: Write infrastructure modules

Data & Machine Learning

data-scientist: Analyze customer behavior dataset
ai-engineer: Build RAG system for document search
mlops-engineer: Set up experiment tracking
ml-engineer: Deploy model to production

Business & Documentation

business-analyst: Create metrics dashboard
docs-architect: Generate technical documentation
api-documenter: Write OpenAPI specifications
content-marketer: Create SEO-optimized content

Multi-Agent Workflows

Subagents coordinate automatically for complex tasks. The system intelligently sequences multiple specialists based on task requirements.

Common Workflow Patterns

Feature Development

"Implement user authentication"
→ backend-architect → frontend-developer → test-automator → security-auditor

Performance Optimization

"Optimize checkout process"
→ performance-engineer → database-optimizer → frontend-developer

Production Incidents

"Debug high memory usage"
→ incident-responder → devops-troubleshooter → error-detective → performance-engineer

Infrastructure Setup

"Set up disaster recovery"
→ database-admin → database-optimizer → terraform-specialist

ML Pipeline Development

"Build ML pipeline with monitoring"
→ mlops-engineer → ml-engineer → data-engineer → performance-engineer

Integration with Claude Code Commands

For sophisticated multi-agent orchestration, use the Claude Code Commands collection which provides 52 pre-built slash commands:

/full-stack-feature   # Coordinates 8+ agents for complete feature development
/incident-response    # Activates incident management workflow
/ml-pipeline         # Sets up end-to-end ML infrastructure
/security-hardening  # Implements security best practices across stack

Subagent Format

Each subagent is defined as a Markdown file with frontmatter:

---
name: subagent-name
description: Activation criteria for this subagent
model: haiku|sonnet|opus  # Optional: Model selection
tools: tool1, tool2       # Optional: Tool restrictions
---

System prompt defining the subagent's expertise and behavior

Model Selection Criteria

• haiku: Simple, deterministic tasks with minimal reasoning
• sonnet: Standard development and engineering tasks
• opus: Complex analysis, architecture, and critical operations

Agent Orchestration Patterns

Sequential Processing

Agents execute in sequence, passing context forward:

backend-architect → frontend-developer → test-automator → security-auditor

Parallel Execution

Multiple agents work simultaneously on different aspects:

performance-engineer + database-optimizer → Merged analysis

Conditional Routing

Dynamic agent selection based on analysis:

debugger → [backend-architect | frontend-developer | devops-troubleshooter]

Validation Pipeline

Primary work followed by specialized review:

payment-integration → security-auditor → Validated implementation

Agent Selection Guide

Architecture & Planning

• backend-architect: API design, microservices, database schemas
• cloud-architect: Infrastructure design, scalability planning
• ui-ux-designer: Interface design, wireframes, design systems

Development by Language

• Systems: c-pro, cpp-pro, rust-pro, golang-pro
• Web: javascript-pro, typescript-pro, python-pro, ruby-pro, php-pro
• Enterprise: java-pro, csharp-pro, scala-pro
• Mobile: ios-developer, flutter-expert, mobile-developer
• Specialized: elixir-pro, unity-developer, minecraft-bukkit-pro

Operations & Infrastructure

• devops-troubleshooter: Production issues, deployment problems
• incident-responder: Critical outages and immediate response
• database-optimizer: Query performance, indexing strategies
• database-admin: Backup, replication, disaster recovery
• terraform-specialist: Infrastructure as Code
• network-engineer: Network debugging, load balancing

Quality & Security

• code-reviewer: Code quality and security analysis
• security-auditor: Vulnerability scanning, compliance
• test-automator: Test suite creation and strategy
• performance-engineer: Application optimization
• debugger: Bug investigation and resolution

Data & Machine Learning

• data-scientist: Data analysis and insights
• ai-engineer: LLM applications and RAG systems
• ml-engineer: Model development and deployment
• mlops-engineer: ML infrastructure and pipelines

Documentation & Business

• docs-architect: Technical documentation generation
• api-documenter: OpenAPI specifications
• business-analyst: Metrics and reporting
• legal-advisor: Legal documentation and compliance

Best Practices

Task Delegation

1. Automatic selection - Let Claude Code analyze context and select optimal agents
2. Clear requirements - Specify constraints, tech stack, and quality standards
3. Trust specialization - Each agent is optimized for their specific domain

Multi-Agent Workflows

1. High-level requests - Allow agents to coordinate complex multi-step tasks
2. Context preservation - Ensure agents have necessary background information
3. Integration review - Verify how different agents' outputs work together

Explicit Control

1. Direct invocation - Specify agents when you need particular expertise
2. Strategic combination - Use multiple specialists for validation
3. Review patterns - Request specific review workflows (e.g., "security-auditor reviews API design")

Performance Optimization

1. Monitor effectiveness - Track which agents work best for your use cases
2. Iterative refinement - Use agent feedback to improve requirements
3. Complexity matching - Align task complexity with agent capabilities

Contributing

To add a new subagent:

1. Create a new .md file with appropriate frontmatter
2. Use lowercase, hyphen-separated naming convention
3. Write clear activation criteria in the description
4. Define comprehensive system prompt with expertise areas

Troubleshooting

Agent Not Activating

• Ensure request clearly indicates the domain
• Be specific about task type and requirements
• Use explicit invocation if automatic selection fails

Unexpected Agent Selection

• Provide more context about tech stack
• Include specific requirements in request
• Use direct agent naming for precise control

Conflicting Recommendations

• Normal behavior - specialists have different priorities
• Request reconciliation between specific agents
• Consider trade-offs based on project requirements

Missing Context

• Include background information in requests
• Reference previous work or patterns
• Provide project-specific constraints

License

MIT License - see LICENSE file for details.

Resources

• Claude Code Documentation
• Subagents Documentation
• Claude Code GitHub
• Claude Code Commands