Letta MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LETTA_API_KEY | Yes | Required API key for Letta Cloud | |
| LETTA_TIMEOUT | No | Timeout in seconds (default: 60) | |
| LETTA_BASE_URL | No | Base URL for Letta API (default: https://api.letta.com) | |
| LETTA_MAX_RETRIES | No | Maximum number of retries (default: 3) | |
| LETTA_DEFAULT_MODEL | No | Default model to use (default: openai/gpt-4o-mini) | |
| LETTA_DEFAULT_EMBEDDING | No | Default embedding model (default: openai/text-embedding-3-small) |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| extensions | {
"io.modelcontextprotocol/ui": {}
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| letta_list_agentsA | List all available Letta agents with pagination support. |
| letta_create_agentB | Create a new Letta agent with memory blocks and tools. |
| letta_get_agentB | Get detailed information about a specific Letta agent. |
| letta_update_agentC | Update an existing agent's configuration. |
| letta_delete_agentA | Delete a Letta agent (requires confirmation). |
| letta_send_messageB | Send a message to a Letta agent and get the response. |
| letta_get_conversation_historyB | Get recent conversation history for a Letta agent. |
| letta_export_conversationC | Export a conversation in various formats. |
| letta_get_memoryC | Get all memory blocks for a Letta agent. |
| letta_update_memoryC | Update a memory block for a Letta agent. |
| letta_create_memory_blockB | Create a new custom memory block for an agent. |
| letta_search_memoryC | Search through agent's conversation memory. |
| letta_list_toolsA | List all available tools in the Letta system. |
| letta_get_agent_toolsC | Get the tools attached to a specific agent. |
| letta_attach_toolB | Attach a tool to an agent. |
| letta_detach_toolB | Detach a tool from an agent. |
| letta_health_checkA | Check the health of the Letta API connection. |
| letta_get_usage_statsC | Get usage statistics for agents. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| get_agents_resource | List of all Letta agents |
| get_tools_resource | List of all available tools |
| get_health_resource | Health status of the Letta connection |
TDQS
Scored across 18 tools
All 18 tools have distinct purposes, covering agent lifecycle, memory, tools, and conversation management. No overlap or ambiguity.
All tools follow the 'letta_verb_noun' pattern, with clear verbs (create, delete, get, list, update, attach, detach, send, export, search) and nouns (agent, memory, tool, conversation).
18 tools is slightly above the ideal range (3-15) but well-justified by the breadth of functionality (agents, memory, tools, conversations, health, usage stats). The count is appropriate for a full agent server.
The tool set covers core operations: CRUD for agents, memory management, tool attachment, conversation interaction, health check, and usage stats. Minor gaps like deleting memory blocks or archiving agents exist but do not hinder typical workflows.