Memory

A personal knowledge graph and memory system for AI assistants using SQLite with optimized text search. Perfect for giving Claude (or any MCP-compatible AI) persistent memory across conversations!

typescript

13

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "spences10-mcp-memory-sqlite": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-memory-sqlite"
      ],
      "env": {
        "SQLITE_DB_PATH": "/path/to/your/sqlite-memory.db"
      }
    }
  }
}

You can give your AI assistant a private memory by storing entities, concepts, and their relationships in a local SQLite database. This MCP server lets you create, search, and relate memory items so your assistant remembers context across conversations without sending data to the cloud.

How to use

Connect your MCP client to the memory server to start building your personal knowledge graph. You can create entities with observations, search for relevant items, read relationships, and establish links between concepts. Use the memory to remember project details, people, and ideas between conversations, and rely on smart search to quickly locate related information.

How to install

Prerequisites: you need Node.js installed on your machine. Use a supported package manager to install the MCP server package.

# Using npm
npm install mcp-memory-sqlite

# Or using pnpm
pnpm add mcp-memory-sqlite
```} ]},{

Configuration and usage notes

Optional: customize where the database file is stored. Set the SQLITE_DB_PATH environment variable to point to your preferred location. If not set, the server uses ./sqlite-memory.db by default.

In Claude Desktop or any MCP-compatible client, configure the memory server to run locally via a standard stdio command. The recommended out-of-the-box configuration runs the package with npx and the appropriate arguments.

Available tools

create_entities

Create or update entities with observations. Accepts an array of entities with a name, entityType, and observations to store new or updated knowledge about a concept.

search_nodes

Search entities and their relations using text search with relevance ranking. Supports case-insensitive and flexible matching across names, types, and observations.

read_graph

Retrieve recently created entities and their relations, returning a snapshot of the memory graph.

create_relations

Establish relationships between entities. Automatically ignores duplicates to keep the graph clean.

delete_entity

Remove an entity and all associated observations and relations from the memory graph.

delete_relation

Delete a specific relation between two entities by source, target, and type.

get_entity_with_relations

Fetch an entity along with all its relations and directly connected entities for in-depth exploration.