Oculairmedia Letta MCP Server

🚀 Letta MCP Server

A Model Context Protocol (MCP) server implementation that interfaces with the Letta API, designed to manage agents, memory blocks, and tools.

🚀 Quick Start

This server offers the following capabilities: creating and managing Letta agents, listing and filtering available agents, creating, reading, updating, and managing memory blocks, listing memory blocks with support for filtering and pagination, attaching memory blocks to agents with custom labels, listing and managing agent tools, and sending messages to agents and receiving responses.

✨ Features

• Create and manage Letta agents.
• List and filter available agents.
• Create, read, update, and manage memory blocks.
• List memory blocks with support for filtering and pagination.
• Attach memory blocks to agents with custom labels.
• List and manage agent tools.
• Send messages to agents and receive responses.

📦 Installation

# Clone the repository
git clone https://github.com/oculairmedia/Letta-MCP-server.git
cd letta-server

# Install dependencies
npm install

# Start the server
node index.js

📚 Documentation

Configuration

Add the following content to the config.json file:

{
  "api_key": "your_api_key",
  "host": "localhost",
  "port": 3000,
  "proxy_enabled": true
}

Tool Instructions

Memory Block Management

• Create a Memory Block:

{
  "success": true,
  "message": "Memory block created successfully.",
  "details": {
    "id": "block-123",
    "content": "Sample memory content."
  }
}

• Read a Memory Block:

{
  "success": true,
  "message": "Memory block retrieved.",
  "data": {
    "id": "block-123",
    "content": "Sample memory content."
  }
}

Agent Management

• Create an Agent:

{
  "success": true,
  "message": "Agent created successfully.",
  "details": {
    "id": "agent-456",
    "name": "Sample Agent"
  }
}

• List Agents:

{
  "success": true,
  "message": "Agents retrieved.",
  "data": [
    {
      "id": "agent-456",
      "name": "Sample Agent"
    },
    // More agents...
  ]
}

Error Handling

All MCP tools return a consistent error format:

{
  "success": false,
  "message": "An error occurred.",
  "error": "Invalid arguments provided.",
  "details": {
    "code": 400,
    "reason": "Bad Request"
  }
}

Performance Considerations

• Memory blocks support pagination for efficient handling of large datasets.
• Tool source code is verified before uploading.
• Support for streaming agent responses to handle long conversations.
• Automatic cleanup of old resources.
• Request rate limiting to prevent API overload.

Contribution Guidelines

1. Fork the repository.
2. Create a feature branch (git checkout -b feature/amazing-feature).
3. Commit your changes (git commit -m 'Add some amazing feature').
4. Push to the branch (git push origin feature/amazing-feature).
5. Submit a Pull Request.

📄 License

This project is licensed under the MIT license. See the LICENSE file for details.