HelloPixDig

Directory › Memory & second brain › agentmemory

agentmemory
Image: official project repository

> GEM MEMORY & SECOND BRAIN · AGENTS & AUTOMATION

agentmemory

by Rohit Ghumare

Long-term memory shared by your agents: they remember what you already explained.

29.3 k stars GitHub Apache-2.0 FreeOpen sourceLocal
View on GitHub →

What it does

  • Works with Claude Code, Codex, Cursor, Gemini CLI, Hermes, OpenClaw and any MCP client
  • One memory shared across several agents, browsable in a local viewer
  • Runs on your machine: your memories stay with you

! This is the memory PixDig uses between Claude Code and Hermes. A memory can hold secrets: don't share it, and back it up.

Install it

Requirements: Docker (easiest) or Node.js 22.

Clone the official repo
git clone https://github.com/rohitg00/agentmemory.git
Follow the README install guide
https://github.com/rohitg00/agentmemory#readme

The prompt for your agent

Paste it into your coding agent: it checks the requirements and installs it for you.

Claude CodeCodexCursorGemini CLIopencode
Prompt
Install agentmemory (https://github.com/rohitg00/agentmemory) on my machine.
1. Check the requirements: Docker (easiest) or Node.js 22. If anything is missing, tell me before installing anything.
2. Follow the official method:
- Clone the official repo: git clone https://github.com/rohitg00/agentmemory.git
- see https://github.com/rohitg00/agentmemory#readme
3. Check that it works, then explain how to launch it and where to start.

Always review what your agent is about to run, especially install commands.