Claude Code Agents

Claude Code Agents

This directory contains specialized AI agents organized by category.

Directory Structure

agents/
├── analysis/                # Code analysis and investigation agents
├── development/            # Code development and refactoring agents
├── documentation/          # Documentation generation agents
├── research/               # Research and information gathering agents
├── security/               # Security analysis agents
├── performance/            # Performance optimization agents
├── README.md               # This file
└── guide.md                # Comprehensive subagent guide

Categories

analysis/

Agents for analyzing and understanding code:

• codebase-locator.md - Find WHERE code lives in a codebase
• codebase-analyzer.md - Understand HOW code works
• code-synthesis-analyzer.md - Analyze recently implemented code changes

development/

Agents for code development tasks:

• tdd-python.md - Test-Driven Development for Python
• code-clarity-refactorer.md - Apply refactoring rules for clarity
• bug-issue-creator.md - Analyze bugs and create GitHub issues

documentation/

Agents for documentation tasks:

• git-diff-documentation-agent.md - Document changes from git diffs
• tech-docs-maintainer.md - Maintain technical documentation
• technical-docs-orchestrator.md - Multi-stage documentation creation
• prompt-engineer.md - Optimize and improve prompts

research/

Agents for research tasks:

• web-docs-researcher.md - Search web for official documentation
• multi-agent-synthesis-orchestrator.md - Orchestrate multiple research agents

security/

Agents for security analysis:

• security-orchestrator.md - Comprehensive security investigations

performance/

Agents for performance optimization:

• memory-profiler.md - Identify memory leaks and optimization opportunities

Using Agents

See guide.md for detailed information about creating, configuring, and using subagents effectively.

Benefits of Subagents

1. Clean Context Isolation - Each subagent starts with a fresh context window
2. Parallel Processing - Multiple subagents can work simultaneously
3. Specialized Expertise - Agents are optimized for specific tasks
4. Consistent Results - Predictable behavior across sessions