Red Team MCP
A multi-agent collaboration platform that provides access to over 1,500 models from 68 providers via the Model Context Protocol. It enables users to assemble and coordinate specialized AI teams using advanced orchestration modes like swarm, debate, and hierarchical workflows.
README
<p align="center"> <img src="src/web/static/logo.svg" alt="Red Team MCP" width="120" height="120"> </p>
<h1 align="center">Red Team MCP</h1>
<p align="center"> <strong>Assemble elite AI agent teams to tackle any challenge</strong> </p>
<p align="center"> <a href="#-quick-start">Quick Start</a> • <a href="#-features">Features</a> • <a href="#-multi-agent-coordination">Multi-Agent</a> • <a href="#-mcp-integration">MCP Integration</a> • <a href="#-api-reference">API</a> </p>
<p align="center"> <img src="https://img.shields.io/badge/providers-68-red" alt="68 Providers"> <img src="https://img.shields.io/badge/models-1500+-red" alt="1500+ Models"> <img src="https://img.shields.io/badge/python-3.10+-blue" alt="Python 3.10+"> <img src="https://img.shields.io/badge/license-AGPL--3.0-blue" alt="AGPL-3.0 License"> </p>
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
<table> <tr> <td width="50%">
🎯 Universal Model Access
- 68 Providers: Anthropic, OpenAI, Google, Groq, Mistral, DeepSeek, and 60+ more
- 1500+ Models: Auto-synced from models.dev
- Unified API: One interface for all providers
</td> <td width="50%">
🤖 Multi-Agent Collaboration
- 5 Coordination Modes: Pipeline, Ensemble, Debate, Swarm, Hierarchical
- Predefined Teams: Writing, Marketing, Research, Technical, Executive
- Custom Teams: Build your own agent configurations
</td> </tr> <tr> <td width="50%">
📡 MCP Integration
- VS Code Ready: Works with GitHub Copilot
- Claude Desktop: Native integration
- Dynamic Tools: All agents exposed as MCP tools
</td> <td width="50%">
🚀 Production Ready
- FastAPI Backend: High-performance async API
- Web Dashboard: HTMX-powered admin interface
- Cost Tracking: Per-request usage analytics
</td> </tr> </table>
🚀 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:
| Mode | Description | Best For |
|---|---|---|
| Pipeline | Agents work sequentially, each building on the previous output | Document workflows, iterative refinement |
| Ensemble | Agents work in parallel, then synthesize results | Comprehensive analysis, multiple perspectives |
| Debate | Agents engage in back-and-forth argumentation | Critical thinking, finding flaws |
| Swarm | CrewAI-powered collaboration with delegation | Complex projects, dynamic task allocation |
| Hierarchical | Manager agent delegates to specialists | Large teams, structured workflows |
Predefined Agent Teams
| Team | Agents | Default Mode |
|---|---|---|
| Writing Team | Creative Writer, Editor, SEO Specialist | Pipeline |
| Marketing Team | Strategist, Brand Manager, Social Media | Hierarchical |
| Research Team | Researcher, Data Scientist, Analyst | Ensemble |
| Technical Team | Expert, Solutions Architect, Security | Debate |
| Executive Team | Strategy, Financial, Operations | Ensemble |
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
| Tool | Description |
|---|---|
list_agents |
List all available agents |
list_teams |
List all predefined teams |
chat |
Chat with a specific agent |
run_team |
Execute a team on a task |
coordinate |
Run multi-agent coordination |
brainstorm |
Generate multiple perspectives |
📖 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
| Endpoint | Method | Description |
|---|---|---|
/api/agents |
GET | List all agents |
/api/teams |
GET | List all teams |
/api/models |
GET | List available models |
/health |
GET | Health check |
/ws/chat |
WS | WebSocket streaming |
🏗️ 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
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing) - Add tests for new functionality
- Ensure all tests pass (
python -m pytest) - 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
<p align="center"> <strong>Ready to assemble your AI team?</strong><br> <a href="#-quick-start">Get Started →</a> </p>
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
E2B
Using MCP to run code via e2b.