Mem0 MCP Server

# Mem0 MCP Server API Reference This document provides a comprehensive reference for all the tools available in the Mem0 MCP Server. ## Basic Memory Operations ### mem0_add_memory ``` mem0_add_memory( text: str, user_id: str, agent_id: str = "", run_id: str = "", metadata: dict = {}, enable_graph: bool = False, includes: str = "", excludes: str = "", timestamp: int = 0, expiration_date: str = "" ) -> dict ``` Adds a memory to Mem0. **Parameters:** - `text` (required): The text content to store as a memory - `user_id` (required): User identifier for the memory - `agent_id` (optional): Agent identifier - `run_id` (optional): Session/conversation identifier - `metadata` (optional): Dictionary of additional metadata - `enable_graph` (optional): Whether to process the memory for knowledge graph - `includes` (optional): Regex pattern for text to include - `excludes` (optional): Regex pattern for text to exclude - `timestamp` (optional): Custom creation time (Unix timestamp) - `expiration_date` (optional): Expiration date in ISO 8601 format **Returns:** Dictionary with status and details of the operation ### mem0_search_memory ``` mem0_search_memory( query: str, user_id: str, agent_id: str = "", run_id: str = "", filters: str = "", threshold: float = 0.0, enable_graph: bool = False, memory_duration: str = "", memory_type: str = "" ) -> dict ``` Searches memories in Mem0. **Parameters:** - `query` (required): The search query - `user_id` (required): User identifier to search within - `agent_id` (optional): Agent identifier to filter by - `run_id` (optional): Session identifier to filter by - `filters` (optional): JSON string of additional filters - `threshold` (optional): Minimum similarity score (0.0-1.0) - `enable_graph` (optional): Whether to use graph for retrieval - `memory_duration` (optional): Filter by "short_term" or "long_term" - `memory_type` (optional): Filter by memory type (conversation, working, attention, episodic, semantic, procedural) **Returns:** Dictionary with search results ### mem0_get_all_memories ``` mem0_get_all_memories( user_id: str, agent_id: str = "", run_id: str = "", limit: int = 0, page: int = 0, page_size: int = 0, enable_graph: bool = False ) -> dict ``` Gets all memories for a user. **Parameters:** - `user_id` (required): User identifier - `agent_id` (optional): Agent identifier to filter by - `run_id` (optional): Session identifier to filter by - `limit` (optional): Maximum number of memories to return - `page` (optional): Page number for pagination - `page_size` (optional): Results per page - `enable_graph` (optional): Whether to include graph context **Returns:** Dictionary with all matching memories ### mem0_get_memory_by_id ``` mem0_get_memory_by_id(memory_id: str) -> dict ``` Gets a specific memory by ID. **Parameters:** - `memory_id` (required): The unique identifier of the memory **Returns:** Dictionary with the memory details ### mem0_delete_memory ``` mem0_delete_memory(memory_id: str) -> dict ``` Deletes a specific memory. **Parameters:** - `memory_id` (required): The unique identifier of the memory to delete **Returns:** Dictionary with status of the operation ### mem0_update_memory ``` mem0_update_memory(memory_id: str, data: str) -> dict ``` Updates a specific memory. **Parameters:** - `memory_id` (required): The unique identifier of the memory to update - `data` (required): The new text content for the memory **Returns:** Dictionary with status and details of the operation ## Specialized Memory Types ### mem0_add_short_term_memory ``` mem0_add_short_term_memory( text: str, user_id: str, run_id: str, memory_type: str = "conversation", metadata: dict = {}, enable_graph: bool = False ) -> dict ``` Adds a short-term memory. **Parameters:** - `text` (required): The text content to store - `user_id` (required): User identifier - `run_id` (required): Session identifier for context - `memory_type` (optional): Type of short-term memory ("conversation", "working", or "attention") - `metadata` (optional): Additional metadata - `enable_graph` (optional): Whether to process for knowledge graph **Returns:** Dictionary with status and details ### mem0_add_episodic_memory ``` mem0_add_episodic_memory( text: str, user_id: str, event_date: str = "", metadata: dict = {}, enable_graph: bool = False ) -> dict ``` Adds an episodic memory (experience or event). **Parameters:** - `text` (required): The text content to store - `user_id` (required): User identifier - `event_date` (optional): Date of the event in ISO format - `metadata` (optional): Additional metadata - `enable_graph` (optional): Whether to process for knowledge graph **Returns:** Dictionary with status and details ### mem0_add_semantic_memory ``` mem0_add_semantic_memory( text: str, user_id: str, category: str = "", metadata: dict = {}, enable_graph: bool = True ) -> dict ``` Adds a semantic memory (factual knowledge). **Parameters:** - `text` (required): The text content to store - `user_id` (required): User identifier - `category` (optional): Category of the knowledge - `metadata` (optional): Additional metadata - `enable_graph` (optional): Whether to process for knowledge graph (default: True) **Returns:** Dictionary with status and details ### mem0_add_procedural_memory ``` mem0_add_procedural_memory( text: str, user_id: str, skill_area: str = "", metadata: dict = {}, enable_graph: bool = False ) -> dict ``` Adds a procedural memory (skills, processes). **Parameters:** - `text` (required): The text content to store - `user_id` (required): User identifier - `skill_area` (optional): Area of skill/expertise - `metadata` (optional): Additional metadata - `enable_graph` (optional): Whether to process for knowledge graph **Returns:** Dictionary with status and details ### mem0_add_memory_selective ``` mem0_add_memory_selective( text: str, user_id: str, includes: str = "", excludes: str = "", metadata: dict = {}, run_id: str = "", enable_graph: bool = False ) -> dict ``` Adds a memory with control over what parts are stored. **Parameters:** - `text` (required): The text content to process - `user_id` (required): User identifier - `includes` (optional): Regex pattern for text to include - `excludes` (optional): Regex pattern for text to exclude - `metadata` (optional): Additional metadata - `run_id` (optional): Session identifier - `enable_graph` (optional): Whether to process for knowledge graph **Returns:** Dictionary with status and details ## Advanced Features ### mem0_update_categories ``` mem0_update_categories( custom_categories: list, explanation: str = "" ) -> dict ``` Updates the project's custom memory categories. **Parameters:** - `custom_categories` (required): List of category dictionaries - `explanation` (optional): Explanation for the update **Returns:** Dictionary with status and details ### mem0_get_categories ``` mem0_get_categories() -> dict ``` Gets the current custom memory categories. **Parameters:** None **Returns:** Dictionary with all configured categories ### mem0_set_instructions ``` mem0_set_instructions( instructions: str, explanation: str = "" ) -> dict ``` Sets custom memory extraction instructions. **Parameters:** - `instructions` (required): The instruction text - `explanation` (optional): Explanation for the update **Returns:** Dictionary with status and details ### mem0_get_instructions ``` mem0_get_instructions() -> dict ``` Gets the current memory extraction instructions. **Parameters:** None **Returns:** Dictionary with the current instructions ### mem0_get_graph_relations ``` mem0_get_graph_relations( user_id: str, entity: str, relation_type: str = "" ) -> dict ``` Gets knowledge graph relationships for an entity. **Parameters:** - `user_id` (required): User identifier - `entity` (required): The entity to get relationships for - `relation_type` (optional): Type of relationship to filter by **Returns:** Dictionary with relationship data ### mem0_send_feedback ``` mem0_send_feedback( memory_id: str, feedback_type: str, comments: str = "" ) -> dict ``` Provides feedback on memory quality. **Parameters:** - `memory_id` (required): The unique identifier of the memory - `feedback_type` (required): Type of feedback (helpful, not_helpful, irrelevant, outdated, incomplete) - `comments` (optional): Additional comments about the feedback **Returns:** Dictionary with status and details