Skip to content
MCP server

ZenBrain Memory

Local SQLite long-term memory for MCP clients, with five memory layers

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, self-hosted
Runs
Locally
Version
0.3.1
Listing updated
3 days ago

About ZenBrain Memory

ZenBrain Memory is an open-source MCP server for knowledge and memory. It runs locally on your machine and is started by your AI client.

ZenBrain gives an MCP client a memory that outlives the conversation, stored in one local SQLite file with no account and no network call. Storing is routed automatically: text that looks like steps becomes a procedure, an emotional weight above 0.5 an episode, a confidence above 0.9 a pinned core memory, and anything else a semantic fact.

Recall searches the episodic, semantic, procedural and core layers, and working memory only when named, returning scored results with the layer each came from. Forgetting removes a memory together with the facts or episodes consolidation linked to it. A consolidation pass turns emotionally weighted recent episodes into facts and decays working-memory slots.

Tools carry MCP annotations, so recall and health are read-only, store only adds, and forget and consolidate are marked destructive. The README calls this an early release whose tool surface may change before 1.0. It needs Node.js 22 or newer and a ZENBRAIN_DB path.

Install

Run in a terminal

claude mcp add zenbrain -- npx -y @zensation/mcp

Add to claude_desktop_config.json

{
  "mcpServers": {
    "zenbrain": {
      "command": "npx",
      "args": [
        "-y",
        "@zensation/mcp"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "zenbrain": {
      "command": "npx",
      "args": [
        "-y",
        "@zensation/mcp"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "zenbrain": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@zensation/mcp"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "zenbrain": {
      "command": "npx",
      "args": [
        "-y",
        "@zensation/mcp"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "zenbrain": {
      "command": "npx",
      "args": [
        "-y",
        "@zensation/mcp"
      ]
    }
  }
}

Configuration

  • ZENBRAIN_DBPath to the SQLite file that holds the memory. A leading ~/ means your home directory; a relative path depends on where the client starts the server (some start it in /), so give an absolute or ~/ path. Use ':memory:' for a store that is discarded when the process exits.
  • ZENBRAIN_CONTEXTSComma-separated context domains, passed to the cross-context layer. No tool uses that layer yet, so the value currently has no effect.

Tools this server provides (5)

Taken from the project documentation; the running server may differ.

zenbrain_store
Write something into long-term memory with automatic layer routing.
zenbrain_recall
Search the long-term layers for what is relevant to a query.
zenbrain_forget
Delete a memory and what consolidation linked to it.
zenbrain_consolidate
Run one maintenance pass that turns weighted episodes into facts.
zenbrain_health
Show how full each memory layer is.

What it does

Layered memory
Working, episodic, semantic, procedural and core layers behind one store.
Automatic routing
Chooses the layer on store from wording, emotional weight and confidence.
Scored recall
Returns results with a 0 to 1 score and the layer of origin.
Linked forgetting
Deleting a memory also removes what consolidation linked to it.
Consolidation pass
Turns weighted episodes into facts and clears decayed working slots.
Tool annotations
Marks read-only, additive and destructive tools for client prompts.

Good for

  • Give Claude Code a memory of project decisions across sessions
  • Keep personal preferences available in Cursor or Claude Desktop
  • Store step-by-step procedures an agent should reuse
  • Correct a wrong memory by forgetting and re-storing it

Pricing

Open source

Apache 2.0 per the site

Prices change. Check zensation.ai before you buy.

ZenBrain Memory FAQ

Does it send data anywhere?

No. It uses one local SQLite file and makes no network call.

Where is the memory stored?

At the path in ZENBRAIN_DB, which defaults to ./zenbrain.db; use an absolute or ~/ path since some clients start servers in /.

What Node version is required?

Node.js 22 or newer.

Is it stable?

The README calls it an early release whose tools may change before 1.0.

Is ZenBrain Memory free?

Yes. ZenBrain Memory is open source, so you can use and self-host it for free.

How do I add ZenBrain Memory to Claude Code?

Run this command in your terminal: claude mcp add zenbrain -- npx -y @zensation/mcp

What are the best alternatives to ZenBrain Memory?

The closest alternatives to ZenBrain Memory are Codebase Memory, XERJ, Claude Code Ultimate Guide, Nautilus Compass Memory and note1.

Alternatives to ZenBrain Memory

  1. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  2. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  3. MCP server for searching the Claude Code Ultimate Guide from any MCP client

    Runs locally
    • Open source
    • 6.5K GitHub stars
    • AI models and LLM tooling
  4. Local-first long-term memory layer for AI agents, served over MCP

    Not checked yet
    • 5 tools
    • Remote
    • Open source
    • 1.3K GitHub stars
    • Knowledge and memory
  5. note1

    MCP server

    MCP server giving AI assistants search and summaries across your note1 meetings

    • 4 tools
    • Remote
    • Free trial
    • 1 GitHub star
    • Productivity