The Memory MCP server gives an agent persistent, knowledge-graph-based memory, so facts, entities, and relationships learned in one session can be recalled in later ones.
Key Features
- Store entities, relations, and observations as a graph
- Recall persisted knowledge across sessions
- Add, update, and query memory through tools
- Simple local persistence
Use Cases
- Give an assistant long-term memory of your preferences
- Build a personal or project knowledge base
- Persist context an agent gathers over time
Getting started
Add Memory MCP to your MCP-capable client — Claude Code, Cursor, Codex, and others — by following the setup in the official repository. Most servers register under your client's mcpServers config with a command, arguments, and any required API keys; Memory MCP documents the exact command and credentials in its README.
