Provides persistent memory capabilities allowing AI agents to maintain and reference information across multiple conversations through a knowledge graph structure.
Memory MCP
A knowledge-graph-based memory system for AI agents that enables persistent information storage between conversations.
Features
- Persistent memory storage using a knowledge graph structure
- Entity-relation model for organizing information
- Tools for adding, searching, and retrieving memories
Tools
The system provides the following MCP tools:
load_knowledge_graph()
: Retrieves the entire knowledge graphget_knowledge_graph_size()
: Returns the current size category of the graph ("small", "medium", or "large")add_entities(entities)
: Adds new entities to the memoryadd_relations(relations)
: Creates relationships between entitiesadd_observations(entity_name, observations)
: Adds observations to existing entitiesdelete_entities(entity_names)
: Removes entities from memorydelete_relations(relations)
: Removes relationshipssearch_nodes(query, search_mode)
: Searches for entities and relations matching a query. Supports three search modes:- "exact_phrase": Matches the entire query as a substring
- "any_token": Matches if any word in the query matches (default)
- "all_tokens": Matches if all words in the query match
open_nodes(names)
: Retrieves specific entities and their relationships between them
Usage
Run the agent with:
The agent will automatically:
- Load its memory at the start of conversations
- Reference relevant information during interactions
- Update its memory with new information when the conversation ends
Exit a conversation by typing q
.
Configuration
Set the memory storage location with the MEMORY_FILE_PATH
environment variable (defaults to memory.json
).
This server cannot be installed
local-only server
The server can only run on the client's local machine because it depends on local resources.
A knowledge-graph-based memory system for AI agents that enables persistent information storage between conversations.
Related MCP Servers
- AsecurityAlicenseAqualityAn improved implementation of persistent memory using a local knowledge graph with a customizable --memory-path. This lets Claude remember information about the user across chats.Last updated -9156247JavaScriptMIT License
- -securityAlicense-qualityA basic implementation of persistent memory using a local knowledge graph. This lets Claude remember information about the user across chats.Last updated -914,70050,763PythonMIT License
- -securityAlicense-qualityA persistent memory implementation using a local knowledge graph that lets Claude remember information about users across conversations.Last updated -4PythonMIT License
- -securityAlicense-qualityEnhanced knowledge graph memory server for AI assistants that uses Neo4j as the backend storage engine, enabling powerful graph queries and efficient storage of user interaction information with full MCP protocol compatibility.Last updated -TypeScriptMIT License