What is the Letta memory MCP server?

It's an MCP server that integrates Letta's memory management system, allowing AI agents to persist and retrieve user context across multiple conversations. You connect it via stdio and authenticate with a Letta API key.

How do I authenticate with the Letta memory server?

Set the LETTA_API_KEY environment variable with your Letta API credentials before running the server. Optionally, provide LETTA_USER_ID to scope memories to specific users.

How do I install and run this MCP server?

Install via npm with `npx @letta-ai/memory-mcp@2.0.2`, then configure your MCP client to use stdio transport with the required LETTA_API_KEY environment variable set.

Can I associate memories with specific users?

Yes, use the LETTA_USER_ID environment variable to tie memories to individual users, enabling personalized context retention across conversations.

What memory operations are available through this server?

The server supports standard memory operations via Letta's API. Refer to the Letta documentation or the server's tool definitions for the complete list of available memory management functions.

MCP ServerSTDIOOfficialv2.0.2

Letta Memory MCP Server

Manage persistent AI memory and context across conversations using Letta's memory backend. For developers building multi-turn AI agents that need to retain and recall user information, preferences, and conversation history.

com.letta/memory-mcp

Try MCP with AI models View source

Hosted URL

Local install

Transport

STDIO

Auth

1 required env var

What the Letta Memory MCP server does

How models use it and what it is built for.

Connect to Letta Memory

Local install — runs as a subprocess.

npx @letta-ai/memory-mcp@2.0.2

Environment variables

Configuration this server reads at startup.

LETTA_API_KEYRequiredSecret
Letta API key for memory operations
LETTA_USER_ID
Unique user identifier for associated memories

Resources

Where to find authoritative docs and source for Letta Memory.

Source repository

Example prompts for Letta Memory

Paste any of these into Agent Studio after connecting Letta Memory.

Store this user preference in memory for future conversations
Retrieve all memories associated with this user ID
Update the conversation context in Letta memory
What memory operations does this MCP server support?

Letta Memory MCP server — FAQ

Common questions about connecting and running Letta Memory.

What is the Letta memory MCP server?
It's an MCP server that integrates Letta's memory management system, allowing AI agents to persist and retrieve user context across multiple conversations. You connect it via stdio and authenticate with a Letta API key.
How do I authenticate with the Letta memory server?
Set the LETTA_API_KEY environment variable with your Letta API credentials before running the server. Optionally, provide LETTA_USER_ID to scope memories to specific users.
How do I install and run this MCP server?
Install via npm with `npx @letta-ai/memory-mcp@2.0.2`, then configure your MCP client to use stdio transport with the required LETTA_API_KEY environment variable set.
Can I associate memories with specific users?
Yes, use the LETTA_USER_ID environment variable to tie memories to individual users, enabling personalized context retention across conversations.
What memory operations are available through this server?
The server supports standard memory operations via Letta's API. Refer to the Letta documentation or the server's tool definitions for the complete list of available memory management functions.

Skip the local setup — run MCP in your browser

MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 60+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.

Open Agent Studio