super-agent-party

Super Agent Party is an AI desktop companion suite that brings together multiple capabilities to automate tasks on your computer. It includes features like a VRM desktop pet, a Task Center for background automation, multi-role group chat with memory, instant messaging integration across popular platforms, live streaming assistants, an AI-powered browser for agents, and an extension system for adding custom capabilities. The server exposes MCP-style interfaces so agents or external tools can connect and control or query the system. Typical usage involves running the server locally (or via Docker) and then using the provided tools or external MCP clients to interact with the agent network, manage extensions, and orchestrate tasks across your computer environment. When running locally, you can access the UI and APIs through the configured port (default 3456) and use the token management endpoint for API keys.

Available tooling includes:

• Quick-start deployment via Docker or source code: you can run a prebuilt image or install from source and run npm scripts to boot the server.
• Extension system: install and create extensions to add new behaviors or integrations for agents.
• API and MCP interfaces: expose OpenAI-compatible interfaces and programmatic MCP endpoints for external connections.
• Access to user-facing dashboards and authentication (token.html) for managing API keys and access.

To interact with the server, connect your MCP client to the host:port where the server is running, and use the provided endpoints to manage agents, tokens, and extensions. If you want to run multiple instances or integrate with other automation tools, leverage the environment variables and the extension framework to tailor capabilities to your environment.

Prerequisites:

• Node.js and npm installed on your machine (for source deployment).
• Docker (optional, for containerized deployment).
• Git (to clone the repository).

Option A: Install from source (Node.js)

1. Install dependencies
   • Clone the repository: git clone https://github.com/heshengtao/super-agent-party.git cd super-agent-party
   • Install Node.js dependencies: npm install
2. Run the server in development mode
   • npm run dev
   • This will start the MCP server and host the UI/API locally (adjust scripts as needed in package.json).

Option B: Run via Docker

1. Pull and run the image docker pull ailm32442/super-agent-party:latest docker run -d -p 3456:3456 -v ${PWD}/super-agent-data:/app/data ailm32442/super-agent-party:latest
2. Access the UI/API at http://localhost:3456/

Option C: Docker Compose (browser-based access)

1. Clone the repo and start services git clone https://github.com/heshengtao/super-agent-party.git cd super-agent-party docker-compose up -d
2. Default credentials: username: root, password: pass (change after first login)
3. Access http://localhost:3456/ and visit http://localhost:3456/token.html for API key management.

Notes:

• If you are building from source, ensure you have network access to install npm dependencies and that your Node.js version matches the project requirements.
• For production deployments, consider configuring a reverse proxy and securing API endpoints.

Tips and common issues:

• If you encounter port conflicts, change PORT in the environment or adjust Docker/compose ports accordingly.
• When using the Docker deployment, the data directory (e.g., ./super-agent-data) is mounted to persist user data and tokens between restarts.
• The extension system is new; check available extensions and ensure compatibility with your MCP client version.
• After first login via Docker Compose, update the default password to ensure security.
• If using source deployment, you may need to expose additional environment variables for integrating with external AI services or APIs. See the README for extension and API documentation.