Letta MCP Server

List all available agents in the Letta system. Use with create_agent to add new ones, get_agent_summary for details, or prompt_agent to interact with them.

Send a message to an agent and get a response. Ensure the agent has necessary tools attached (see attach_tool) first. Use list_agents to find agent IDs.

List all tools available for a specific agent. Use attach_tool to add more tools or list_mcp_tools_by_server to discover available tools.

Create a new Letta agent with specified configuration. After creation, use attach_tool to add capabilities, attach_memory_block to configure memory, or prompt_agent to start conversations.

Attach one or more tools (by ID or name) to an agent. If a name corresponds to an MCP tool not yet in Letta, it will be registered first. Find tools with list_mcp_tools_by_server or create custom ones with upload_tool. Use list_agent_tools to verify attachment.

List all memory blocks available in the Letta system. Use create_memory_block to add new ones, update_memory_block to modify, or attach_memory_block to link them to agents.

Get full details of a specific memory block by ID. Use list_memory_blocks to find block IDs. After reading, use update_memory_block to modify content.

Update the contents and metadata of a memory block. Use list_memory_blocks to find block IDs, or read_memory_block to see current content before updating.

Attach a memory block to an agent. Use list_memory_blocks to find blocks, create_memory_block to make new ones. Common labels: "persona", "human", "system".

Create a new memory block in the Letta system. Common labels: "persona", "human", "system". Use attach_memory_block to link to agents, or update_memory_block to modify later.

Delete a memory block by ID. Use list_memory_blocks to find block IDs. WARNING: This action is permanent and cannot be undone.

Search an agent's memory across both archival passages and conversation messages. Supports text search with optional date filtering. Use source parameter to search only archival or messages for better performance.

Upload a new tool to the Letta system. Use with attach_tool to add it to agents, or list_agent_tools to verify attachment.

List all available tools for a specific MCP server. Use list_mcp_servers first to see available servers, then add_mcp_tool_to_letta to import tools into Letta.

List all configured MCP servers on the Letta server. Use with list_mcp_tools_by_server to explore available tools from each server.

Get the full state of a specific agent by ID. Similar to get_agent_summary but returns complete details. Use list_agents to find agent IDs.

Update an existing agent by ID with provided data. Use get_agent_summary to see current config, list_llm_models/list_embedding_models for model options. For tools, use attach_tool instead.

Delete a specific agent by ID. Use list_agents to find agent IDs. For bulk deletion, use bulk_delete_agents. WARNING: This action is permanent.

List available LLM models configured on the Letta server. Use with create_agent or modify_agent to set agent model preferences.

List available embedding models configured on the Letta server. Use with create_agent or modify_agent to set agent embedding preferences.

Retrieve the memories in an agent's archival memory store (paginated query). Use create_passage to add new memories, modify_passage to edit, or delete_passage to remove them.

Insert a memory into an agent's archival memory store. Use list_passages to view existing memories, modify_passage to edit, or delete_passage to remove.

Modify a memory in the agent's archival memory store. Use list_passages to find memory IDs. Currently only supports updating the text content.

Delete a memory from an agent's archival memory store. Use list_passages to find memory IDs. WARNING: This action is permanent.

Search an agent's archival memory using semantic similarity. Returns passages most similar to the query. Use list_passages for text-based search or pagination, create_passage to add memories.

Export an agent's configuration to a JSON file and optionally upload it. Use import_agent to recreate the agent later, or clone_agent for a quick copy. Use list_agents to find agent IDs.

Import a serialized agent JSON file and recreate the agent in the system. Use export_agent to create the JSON file, then modify_agent or attach_tool to customize the imported agent.

Creates a new agent by cloning the configuration of an existing agent. Use list_agents to find source agent ID. Alternative to export_agent + import_agent workflow. Modify the clone with modify_agent afterwards.

Attaches a specified tool to multiple agents based on filter criteria (name or tags). Use list_agents to find agents and list_mcp_tools_by_server or upload_tool to get tool IDs.

Provides a concise summary of an agent's configuration, including core memory snippets and attached tool/source names. Use list_agents to find agent IDs. Follow up with modify_agent to change settings or attach_tool to add capabilities.

Deletes multiple agents based on filter criteria (name or tags) or a specific list of IDs. Use list_agents first to identify agents to delete. WARNING: This action is permanent.

Registers a tool from a connected MCP server as a native Letta tool AND attaches it to a specified agent. Use list_mcp_tools_by_server to find available tools, and list_agents to get agent IDs.

List all available prompt templates including wizards and workflows

Execute a registered prompt template. Use this to run wizards, workflows, and guided interactions.

Retrieve messages from an agent's conversation history. Returns paginated message history including user messages, assistant responses, tool calls, and system messages. Use for reviewing past conversations or debugging agent behavior.

Store a conversation entry in an agent's archival memory. Use this to record conversations from external sources (like Claude Code sessions) that should be searchable via search_archival_memory. Entries are formatted with metadata for easy filtering.