Letta MCP Server

Name: Letta MCP Server
Author: oculairmedia

@oculairmedia

Repository

About Letta MCP Server

MCP server to manage letta server and comunicate with agents

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "Letta-MCP-server": {
      "command": "docker",
      "args": [
        "pull",
        "ghcr.io/oculairmedia/letta-mcp-server-rust:rust-latest"
      ]
    }
  }
}

Tools

Overview

What is Letta MCP Server?

A high-performance Model Context Protocol (MCP) server for Letta AI, built with Rust and the TurboMCP framework, providing 7 consolidated tools covering 103 operations for managing agents, memory, tools, sources, jobs, files/folders, and MCP servers.

How to use Letta MCP Server?

Install globally via npm (npm install -g letta-mcp-server) or use Docker. Configure required environment variables LETTA_BASE_URL and LETTA_PASSWORD. Choose transport mode via TRANSPORT (stdio or http). Launch with stdio for Claude Desktop/Cursor or HTTP for production deployments.

Key features of Letta MCP Server

7 consolidated tools covering 103 operations via discriminator pattern
High performance: Rust, ~10-30MB memory, <500ms startup
Dual transport: stdio (Claude Desktop, Cursor) and HTTP (production)
Response size optimizations with 68–96% reduction for LLM context
Type-narrowed schemas with per-parameter descriptions
Multi-platform support for macOS, Linux, Windows (x64 and arm64)
Compile-time routing with zero runtime overhead

Use cases of Letta MCP Server

Manage agent lifecycle, messaging, conversations, and context export/import
Unify memory operations: core memory, blocks, archival passages, archives, and search
Administer tools: CRUD, attach/detach, bulk operations, and schema generation
Handle data sources: upload files, manage passages, and attach to agents
Monitor jobs: track active jobs, cancel, and list all job statuses

FAQ from Letta MCP Server

What dependencies does Letta MCP Server require?

It requires a running Letta AI instance (0.16.x compatible) accessible via the LETTA_BASE_URL and authenticated with LETTA_PASSWORD. The server binary is self-contained when installed via npm or Docker.