Red Team MCP

Assemble elite AI agent teams to tackle any challenge

Quick Start • Features • Multi-Agent • MCP Integration • API

Red Team MCP is a multi-agent collaboration platform that connects to 68 providers and 1500+ models via models.dev. Build specialized agent teams, coordinate complex workflows, and integrate seamlessly with VS Code and Claude Desktop through the Model Context Protocol (MCP).

✨ Features

🚀 Quick Start

Option A: Docker (Recommended)

git clone https://github.com/yourusername/red-team-mcp.git
cd red-team-mcp
cp .env.example .env
# Edit .env with your API keys

docker compose up -d
# Open http://localhost:8000/ui/

Option B: Local Install

git clone https://github.com/yourusername/red-team-mcp.git
cd red-team-mcp
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate
pip install -r requirements.txt

Configure API Keys

cp .env.example .env
# Edit .env with your API keys

Run

# Start the web server & dashboard
python main.py serve
# Open http://localhost:8000/ui/

# Or use the CLI
python main.py chat "What is machine learning?"

# Or start the MCP server
python main.py mcp

🤝 Multi-Agent Coordination

Red Team MCP excels at coordinating multiple AI agents on complex tasks. Choose from 5 coordination modes:

Predefined Agent Teams

Example: Multi-Agent Request

curl -X POST "http://localhost:8000/api/multi-agent" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "Analyze the competitive landscape for AI startups",
    "coordination_mode": "ensemble",
    "agents": ["financial_analyst", "strategy_consultant", "technical_expert"]
  }'

📡 MCP Integration

Red Team MCP provides a Model Context Protocol server for seamless integration with AI assistants.

VS Code Setup

Create .vscode/mcp.json in your project:

{
  "servers": {
    "red-team-mcp": {
      "command": "python",
      "args": ["-m", "src.mcp_server_dynamic"],
      "cwd": "/path/to/red-team-mcp"
    }
  }
}

Claude Desktop Setup

Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "red-team-mcp": {
      "command": "python",
      "args": ["/path/to/red-team-mcp/main.py", "mcp"]
    }
  }
}

Available MCP Tools

📖 API Reference

Chat Endpoint

POST /api/chat
Content-Type: application/json

{
  "agent_id": "creative_writer",
  "message": "Write a tagline for an AI product",
  "temperature": 0.8,
  "max_tokens": 500
}

Multi-Agent Endpoint

POST /api/multi-agent
Content-Type: application/json

{
  "query": "Develop a go-to-market strategy",
  "coordination_mode": "hierarchical",
  "agents": ["marketing_strategist", "sales_analyst"],
  "rebuttal_limit": 3
}

Run Team Endpoint

POST /api/team/{team_id}/run
Content-Type: application/json

{
  "query": "Create a blog post about AI trends",
  "coordination_mode": "pipeline"
}

Additional Endpoints

🏗️ Architecture

red-team-mcp/
├── main.py                    # CLI entry point
├── config/config.yaml         # Configuration
├── src/
│   ├── api/                   # FastAPI application
│   │   ├── app.py            # App factory
│   │   ├── endpoints.py      # REST endpoints
│   │   └── websockets.py     # WebSocket handlers
│   ├── agents/               # Agent implementations
│   │   ├── configurable_agent.py
│   │   └── coordinator.py    # Multi-agent coordination
│   ├── web/                  # Dashboard UI
│   │   ├── routes.py
│   │   └── templates/        # HTMX templates
│   ├── providers/            # 68 provider implementations
│   ├── config.py             # Configuration management
│   ├── models.py             # Model selector
│   ├── db.py                 # SQLite persistence
│   └── mcp_server_dynamic.py # MCP server
└── mcp_servers/              # Generated MCP servers

⚙️ Configuration

Environment Variables

# Core providers
ANTHROPIC_API_KEY=your_key
OPENAI_API_KEY=your_key
GOOGLE_API_KEY=your_key
GROQ_API_KEY=your_key
DEEPSEEK_API_KEY=your_key

# And 60+ more providers supported!

config.yaml

api:
  host: "0.0.0.0"
  port: 8000
  rate_limit: "100/minute"

models:
  default: "claude-sonnet-4-20250514"

agents:
  predefined:
    - id: my_custom_agent
      name: Custom Agent
      model_id: gpt-4o
      provider: openai
      role: Specialist
      goal: Help with specific tasks

🧪 Development

# Run tests
python -m pytest tests/ -v

# Run with hot reload
python main.py serve --reload

# Generate MCP servers
python main.py generate-mcp --all

📊 Web Dashboard

Access the admin dashboard at http://localhost:8000/ui/ to:

• 💬 Chat with any agent interactively
• 👥 Manage Teams and agent configurations
• 📈 View Statistics on usage and costs
• ⚙️ Configure providers and settings
• 📤 Export configurations

🤝 Contributing

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing)
3. Add tests for new functionality
4. Ensure all tests pass (python -m pytest)
5. Submit a pull request

📄 License

AGPL-3.0 License - see LICENSE for details.

🙏 Acknowledgments

• models.dev - Comprehensive model database
• CrewAI - Agent orchestration framework
• FastAPI - High-performance web framework
• All 68 AI providers making their models accessible

Ready to assemble your AI team?
Get Started →