π€ A unified AI agent orchestration system featuring 10's of specialized agents across 14 categories for modern software development. Built with advanced coordination intelligence, resilience engineering, and structured logging capabilities.
AI Agentsagentagentic-aiai-agentsclaudeclaude-code
Installation
# Add to your Claude Code skills
git clone https://github.com/avivl/claude-007-agents
README.md
π Claude Code: Comprehensive AI Agent System for Developers
Enhance your development workflow with specialized AI agents that help deliver quality code through orchestration, resilience engineering, and organizational memory.
Open-source AI agents for enhanced development workflows
π Why Claude Code Agents Can Improve Your Development
β‘ Specialized Expertise Across Technologies
Reduce context switching between documentation, Stack Overflow, and your IDE. Get access to specialized agents covering major frameworks, languages, and domains.
# Instead of spending hours researching...
claude "Build a resilient authentication system with circuit breakers and structured logging"
# β @rails-expert + @resilience-engineer + @security-specialist work in perfect coordination
π§ AI System with Organizational Memory
These agents can remember and learn from projects when configured with Basic Memory MCP, helping build institutional knowledge over time.
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
151,568
claudecode-subagents
sub-agents
subagents
Pattern Reuse: "Use patterns from previous authentication implementations"
Context Building: "Reference similar microservice architectures we've built"
Knowledge Evolution: "Update our React component patterns based on this success"
π Agent Orchestration System
The system includes coordination features that can help with agent selection and workflow management for complex tasks.
# Just describe what you want - the system handles everything automatically:
claude "Build an e-commerce platform"
# β Vibe Coding Coordinator: 15-20 minute preparation phase, analyzes requirements
# β Exponential Planner: Strategic planning with capability scaling awareness
# β Parallel Coordinator: Coordinates (@react-expert + @rails-expert + @payment-specialist + @security-specialist)
# β Session Manager: Maintains context across extended development sessions
# β Safety Specialists: Pre-deployment validation and architectural analysis
π IMPORTANT: Bootstrap Required β οΈ
β‘ ONE-TIME SETUP: Every project needs bootstrapping before using Claude 007 Agents!
# After installing agents, bootstrap each project once:
claude "Use @bootstrap-orchestrator to analyze and setup this project"
Why Bootstrap?
π Analyzes your tech stack and selects optimal agents
βοΈ Creates proper CLAUDE.md configuration with commit attribution
π― Sets up project-specific agent selection and workflows
π Configures Task Master integration for codebase-aware development
β Validates complete system readiness
Skip this step and agents won't work properly in your project!
Sequential Thinking MCP: Complex reasoning with adaptive planning and course correction
5. π Professional-Grade Workflows
Vibe Coding: 15-20 minute autonomous development preparation phases
Parallel Execution: Coordinated multi-agent development with tool call batching
Exponential Planning: Long-term development planning with AI capability scaling awareness
Sequential Reasoning: Multi-step problem analysis with adaptive strategy revision
Safety Validation: Pre-deployment testing and architectural safety analysis
Dynamic Permissions: Real-time permission management with risk-based escalation
Automatic Commit Attribution: Every commit traces back to contributing agents
Quality Gates: Pre-commit hooks with comprehensive linting
6. π§ Organizational Intelligence
Cross-Project Learning: Patterns and knowledge shared across all projects
Decision History: Track architectural decisions and their outcomes
Performance Analytics: Agent effectiveness measurement and optimization
Knowledge Graphs: Dynamic project understanding with real-time updates
π― Perfect For
| Team Size | Use Case | Key Benefits |
|---------------|--------------|-----------------|
| Solo Developers | Full-stack development, learning new technologies | Instant expertise, quality assurance, best practices |
| Small Teams (2-10) | Startup MVPs, rapid prototyping, scaling challenges | Coordinated development, knowledge sharing, quality consistency |
| Enterprise Teams | Complex systems, compliance, technical debt | Orchestrated workflows, organizational memory, standardized practices |
| Open Source | Community projects, documentation, code reviews | Automated quality, contributor onboarding, pattern consistency |
β‘ Get Started in 60 Seconds
π BOOTSTRAP FIRST: One-Time Project Setup β
IMPORTANT: You must bootstrap the agent system once for every new repo/project before using Claude 007 Agents.
# 1. Clone the system
git clone https://github.com/avivl/claude-007-agents.git
cd claude-007-agents
# 2. Copy to your project (any project)
cp -r .claude/agents /path/to/your/project/.claude/
cp agents.json /path/to/your/project/
# 3. BOOTSTRAP your project (REQUIRED - once per project)
cd /path/to/your/project
claude "Use @bootstrap-orchestrator to analyze and setup this project"
# 4. Start development immediately
claude "Use @software-engineering-expert to begin development"
What Bootstrap Does:
π Analyzes your codebase (tech stack, architecture, complexity)
π― Selects optimal agents based on your technology stack
βοΈ Creates CLAUDE.md with proper agent configuration and commit attribution
π Sets up Task Master integration (if desired)
π Configures MCP servers for enhanced capabilities
β Validates complete system readiness
Bootstrap Works With:
β New projects (empty directories) β Full system initialization
β Existing projects without Claude β Intelligent integration
β Projects with existing CLAUDE.md β Enhancement and upgrade
β Partial setups β Completion of missing components
Prerequisites & Installation
Required Software:
# Essential: Claude Code CLI (required)
# Visit: https://docs.anthropic.com/en/docs/claude-code
# Essential: Node.js for bootstrap system and advanced features
node --version # v18+ REQUIRED for bootstrap engine and Task Master integration
npm --version # npm v6+ required for package management
Required Packages (for enhanced features):
# Task Master integration (for codebase-aware development)
npm install -g task-master-ai
# Basic Memory MCP (for organizational learning)
npm install -g @modelcontextprotocol/server-memory
# Context7 MCP (for live documentation)
npm install -g @upstash/context7-mcp
# Bootstrap System Dependencies (automatically handled)
# These are included when you clone the repo - no separate installation needed
Environment Setup:
# API keys (add to your ~/.bashrc or ~/.zshrc)
export ANTHROPIC_API_KEY="your_anthropic_key_here"
export PERPLEXITY_API_KEY="your_perplexity_key_here" # For research features
export OPENAI_API_KEY="your_openai_key_here" # Optional
export GOOGLE_API_KEY="your_gemini_key_here" # Optional (for Zen MCP)
# Or create .env file in your project directory
echo "ANTHROPIC_API_KEY=your_key_here" > .env
echo "PERPLEXITY_API_KEY=your_key_here" >> .env
