Local Memory MCP Server

Ever wanted the ChatGPT memory feature but **across all your LLMs** and stored on **your own hardware**? Ever hate how there's a **limit to how many

Local serverstdioPython

What is the Local Memory MCP MCP server?

Local Memory MCP MCP server is a locally run integration for AI assistants that speak the Model Context Protocol. Ever wanted the ChatGPT memory feature but across all your LLMs and stored on your own hardware? Ever hate how there's a limit to how many memories ChatGPT can store, and that you can't segment your memories into different.

What you get

  • 🧠 Remembers across sessions - User preferences, project details, conversation history
  • 🎯 Finds relevant context - Semantic search surfaces the right memories at the right time
  • 🏢 Organizes by domain - Separate contexts for work, health, personal life (PostgreSQL)
  • 🔒 Stays private - All data stored locally under your control
  • Works immediately - Drop-in compatibility with Claude Desktop and MCP clients

What the assistant can call

Once Local Memory MCP is connected, these are the calls the assistant has available:

  • store_memory — Store new information in persistent memory with automatic semantic indexing. - SQLite: store_memory(content, source?, importance?) -
  • update_memory — Modify existing memories while preserving search indexing. - SQLite: update_memory(memory_id, content?, importance?) - PostgreSQL
  • search_memories — Find relevant memories using semantic or keyword search. - SQLite: search_memories(query, limit?, use_vector?) - PostgreSQL
  • summarize_memories — Generate intelligent summaries of retrieved memory collections. - Input: List of memory objects - Output: Structured summary highlighting key
  • Prerequisites — The Prerequisites tool exposed by this server

Configuration and credentials

You will need 2 environment variables: OLLAMA_API_URL, POSTGRES_HOST. The server will not start without them, which is usually why the tools fail to appear on a first run. Keep credentials in your client's env block or a secrets manager rather than in a file you might commit.

  • Docker (for containerized deployment) or Python 3.12+ (for local installation) - Ollama with nomic-embed-text model (optional but recommended for enhanced semantic search)

Setting it up

cunicopia/local-memory-mcp on a container image is all you need. Most clients run it directly, so configuration is a few lines and a restart.

Choosing this one

Among the database access options, the useful question is rarely "what can it do" but "what does it cost you to run" — permissions, credentials, and how much of your context its toolset consumes. Local Memory MCP's toolset — store_memory, update_memory, search_memories and 2 more — is a fair guide to whether it matches your workflow. It is maintained by cunicopia-dev; worth a glance at recent repository activity before you build anything load-bearing on it.

SyncDev reviews every entry in this directory against the project's own documentation before publishing, and revisits them as servers change.

Before you rely on it

  • It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
  • Missing credentials fail quietly in some clients — if no tools show up, check the environment block first.
  • MCP clients confirm each tool call by default. Leave that on until you have watched what the local memory mcp mcp server does with a few real requests.

Available tools

ToolWhat it does
store_memoryStore new information in persistent memory with automatic semantic indexing. - **SQLite**: store_memory(content, source?, importance?) - **PostgreSQL**: store_memory(content, domain?, source?, importance?) - **Examples**
update_memoryModify existing memories while preserving search indexing. - **SQLite**: update_memory(memory_id, content?, importance?) - **PostgreSQL**: update_memory(memory_id, content?, importance?, domain?) - **Use case**: Update o
search_memoriesFind relevant memories using semantic or keyword search. - **SQLite**: search_memories(query, limit?, use_vector?) - **PostgreSQL**: search_memories(query, domain?, limit?) - **Examples**: - "What programming languages d
summarize_memoriesGenerate intelligent summaries of retrieved memory collections. - **Input**: List of memory objects - **Output**: Structured summary highlighting key patterns and insights - **Use case**: Create context summaries for com
PrerequisitesThe Prerequisites tool exposed by this server.

How to install the Local Memory MCP MCP server

{
  "mcpServers": {
    "local-memory": {
      "command": "docker",
      "args": ["run", "-i", "--rm", "cunicopia/local-memory-mcp"],
      "env": {
        "OLLAMA_API_URL": "your-value",
        "POSTGRES_HOST": "your-value"
      }
    }
  }
}

Add to claude_desktop_config.json, then restart Claude Desktop.

Configuration

  • Docker (for containerized deployment) or Python 3.12+ (for local installation) - Ollama with nomic-embed-text model (optional but recommended for enhanced semantic search)
VariableDescriptionRequired
OLLAMA_API_URLEndpoint or connection string the server talks to.Yes
POSTGRES_HOSTEndpoint or connection string the server talks to.Optional

Example prompts to try

  • Use Local Memory MCP to store memory.
  • Use Local Memory MCP to update memory.
  • Use Local Memory MCP to search memories.

Frequently asked questions

It connects Local Memory MCP to MCP-compatible AI assistants such as Claude and Cursor, exposing 5 tools (store_memory, update_memory, search_memories, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Local Memory MCP directly.