Skip to content
MCP server

Counterparts

Local-first memory for Claude Code that scores, fades and consolidates what it keeps

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, macOS, Linux
Runs
Locally
Version
0.3.47
Listing updated
3 days ago

About Counterparts

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

Counterparts models memory on human recall. Each memory gets an importance score when written, set by the AI that lived the conversation. Memories written with expressed feeling start stronger and fade more slowly, unused ones fade over days you actually worked together, and very faint ones are filed away instead of deleted. Core memories about identity do not fade.

Recalling a memory strengthens it and pulls linked ones along, and a memory with a date attached surfaces on that day unprompted. Once a day it consolidates, merging similar memories. The site marks interference between old and new memories as still in development.

It stores data on your machine, by default under ~/.counterparts/store, needs no account or API key, and is MIT-licensed. Claude Code is the supported host, with a Claude Desktop install option on macOS.

Install

Run in a terminal

claude mcp add memory -- npx -y counterparts

Add to claude_desktop_config.json

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

Add to ~/.cursor/mcp.json

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

Add to .vscode/mcp.json

{
  "servers": {
    "memory": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "counterparts"
      ]
    }
  }
}

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

{
  "mcpServers": {
    "memory": {
      "command": "npx",
      "args": [
        "-y",
        "counterparts"
      ]
    }
  }
}

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

{
  "mcpServers": {
    "memory": {
      "command": "npx",
      "args": [
        "-y",
        "counterparts"
      ]
    }
  }
}

Configuration

  • COUNTERPARTS_DATA_DIRWhere the memory lives. Default: ~/.counterparts/store.
  • COUNTERPARTS_CONFIGThe host configuration file. Default: ~/.counterparts/claude-code.json.
  • COUNTERPARTS_UPDATE_CHECKSet to off to stop the daily npm version check.

What it does

Importance scoring
The AI rates each memory when written.
Decay and forgetting
Unused memories fade; core ones persist.
Association
Memories used together get linked.
Prospective recall
Dated memories return on the right day.
Daily consolidation
Similar memories are merged offline.
Local storage
Data stays on your machine.

Good for

  • Continue a project across Claude sessions
  • Remember decisions and their reasons
  • Resurface a dated reminder automatically
  • Keep memory private with no cloud account

Pricing

Open source

MIT license, runs locally

Prices change. Check counterparts.ai before you buy.

Counterparts FAQ

Where is memory stored?

On your machine, by default in ~/.counterparts/store.

Does it need an account?

No API keys and no account.

Which clients?

Claude Code, with a Claude Desktop install on macOS.

How is it licensed?

MIT.

Is Counterparts free?

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

How do I add Counterparts to Claude Code?

Run this command in your terminal: claude mcp add memory -- npx -y counterparts

What are the best alternatives to Counterparts?

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

Alternatives to Counterparts

  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