agent-memory
by tigerless-labs
agent-memory is a long-term memory system for AI agents that stores what they learn as plain Markdown files, with a local search index beside them. It lets tools such as Claude Code and Codex share one store of memories between sessions without needing an API key.
- Stars
- 4.7k +84 a day, est.
- Licence
- MIT
- Runs on
- Your computer
- Setup
- Moderate
- Language
- Python
- Age
- 56 days
What you can do with it
- Give Claude Code and Codex one shared memory store that persists between sessions
- Search past decisions and notes with ranked local recall, then open only the files you need
- Review and tidy memories with a background consolidation pass, confirming any deletions yourself
Make money with it
A freelance developer could offer agent-assisted build and maintenance work for clients and use persistent memory to keep long projects consistent.
An idea, not financial advice. Check the licence before charging for anything built on it.
Ways a business could use it
- Keep an AI coding assistant aware of past project decisions, so a freelancer spends less time re-explaining context to it each session
- Maintain a shared, searchable record of client preferences and technical choices that several agents can use
Can you use it commercially?
Free to use, change and sell, including in paid products. Keep the copyright notice.
Not legal advice. Read the project's licence file before building a product on it.
Getting it running
It is a Python command-line tool (the `mem` command) that works with agents able to run shell commands; the excerpt of the README does not show exact install steps.
Follow the official instructions in the project README.
Things to check
- Requires a host agent such as Claude Code or Codex, which may need its own paid subscription
- Benchmark results are the project's own and the README excerpt is truncated
- Install steps were not visible in the supplied README
Checked 27 Oct 2026. Summary drafted by AI from the project's README.
