Skip to content
Get Friday PicksSubmit
SkillClaude Code logoClaude CodeCodex logoCodexCoding Agent SkillsVerified

ai-memory: one memory and handoff for all your coding agents

A small server you run yourself that Claude Code, Codex, Cursor, Gemini CLI and 20+ other agents share. It records sessions quietly, keeps what it learned as plain markdown in git, and hands off the work when you switch tools: quit Claude Code mid-task, open Codex in the same folder, and it knows where you stopped, what failed and what is still open.

In short: A small server you run yourself that Claude Code, Codex, Cursor, Gemini CLI and 20+ other agents share. It records sessions quietly, keeps what it learned as plain markdown in git, and hands off the work when you switch tools: quit Claude Code mid-task, open Codex in the same folder, and it knows where you stopped, what failed and what is still open. It is a free skill for people who use Claude Code or Codex and want help with coding agent skills.

Checked on · Added

Open on GitHub · ★ 14.2k

Works with

Agent / toolAccess / costLink
Claude CodePaid Claude plan or APIclaude.com/claude-code
CodexChatGPT plan or APIopenai.com/codex
GitHub repoOpen source · ★ 14.2kakitaonrails/ai-memory

Try it in 60 seconds

curl -fsSL https://github.com/akitaonrails/ai-memory/releases/latest/download/install.sh -o install.sh   # read it, then: sh install.sh
ai-memory setup   # shows the plan and asks before wiring each agent
claude plugin marketplace add akitaonrails/ai-memory && claude plugin install ai-memory@ai-memory

Install command copied from the project README. Check the README for the latest version.

What it does

A small server you run yourself that Claude Code, Codex, Cursor, Gemini CLI and 20+ other agents share. It records sessions quietly, keeps what it learned as plain markdown in git, and hands off the work when you switch tools: quit Claude Code mid-task, open Codex in the same folder, and it knows where you stopped, what failed and what is still open.

ai-memory runs a small server on your own machine (or a home server or team box) that every coding agent you use reads from and writes to. Hooks quietly record what happened in each session, strip private data, and turn…

Short excerpt from the README of akitaonrails/ai-memory. Read the full README on GitHub.

Inside the repo

Why ai-memoryHow it worksSupport matrixComing from another tool?Quick startLinux and macOS (two commands, no root)Windows (two commands, no admin)Arch Linux (AUR)Fedora (RPM)macOS (menu bar app)DockerNixOS

A prompt to try

Before you start, check ai-memory for the last handoff on this repo and continue from the open questions.

Written by us so you can reuse it. Replace each [bracket] with your details.

Step by step

  1. Open the GitHub repo and read Quick start and Security.
  2. Download install.sh from the latest release, read it, then run it (it checks the release checksum).
  3. Run ai-memory setup: it shows which agents it found and asks before wiring each one.
  4. Restart your agents and ask one: what do you remember about this project?

Watch out for

It records your prompts and tool calls (filtered for secrets before storage) and installs a background service plus hooks in each agent, so run ai-memory setup --plan first to see what it will change. The Docker quick start has no login, so keep it bound to your own machine. Our scan flags were read by hand: they are uninstall lines in the docs, a Keychain note for test speed and an old curl-into-shell line in the changelog; download and read the installer instead of piping it.

Docs, demos & materials

Common questions

What does ai-memory do?

A small server you run yourself that Claude Code, Codex, Cursor, Gemini CLI and 20+ other agents share. It records sessions quietly, keeps what it learned as plain markdown in git, and hands off the work when you switch tools: quit Claude Code mid-task, open Codex in the same folder, and it knows where you stopped, what failed and what is still open.

Which AI agents can use ai-memory?

It works with Claude Code and Codex. The code is on GitHub at akitaonrails/ai-memory.

Is ai-memory free?

Free & open source (MIT). Runs on your own computer or server (Linux, macOS, Windows, Docker); no API key needed. The optional note-merging pass uses an LLM only if you set one up.

How do I install ai-memory?

Run the install command from the README: “curl -fsSL https://github.com/akitaonrails/ai-memory/releases/latest/download/install.sh -o install.sh # read it, then: sh install.sh”. Check the README on GitHub for the latest version, then follow the steps on this page.

Is ai-memory safe to install?

We read 446 instruction and code files in the repo and found no red flags. That is not a guarantee, so review anything that runs commands on your machine.

Sources & credit

All credit to the original authors. We rewrite in our own words and link back. Something wrong? Report it.

Related skills and use cases in Coding Agent Skills

See all Coding Agent Skills entries →

Related

CleanYourTV: your agent removes the ads from your TV thumbnailSKILL

CleanYourTV: your agent removes the ads from your TV

Your coding agent connects to an Android or Google TV over Wi-Fi, switches off the ad and tracking apps, and puts in a clean, Apple TV-style home screen. Nothing is uninstalled.

Claude Code logoCodex logoGemini logoClaude Code +2★ 14.8k
unslop: turn AI-looking UI into a designed interface thumbnailSKILL

unslop: turn AI-looking UI into a designed interface

A Claude skill that reviews a vibecoded screen like a designer: it lists the tell-tale AI touches, proposes a direction from your own product, then fixes spacing, colour, icons, copy, type and layout pass by pass. Works on dashboards, apps, landing pages, or a live site from a link.

Claude Code logoClaude Code★ 540
ECC: a full work system for your coding agent thumbnailSKILL

ECC: a full work system for your coding agent

One install gives your agent a plan, test, build, review and remember routine, with 71 helper agents, about 300 skills, memory between sessions and a built-in scanner for risky agent settings.

Claude Code logoCodex logoCursor logoClaude Code +3★ 285.3k
Superpowers: plan, test and debug before coding thumbnailSKILL

Superpowers: plan, test and debug before coding

A skills framework that makes agents brainstorm, plan, test and debug systematically.

Claude Code logoCodex logoClaude Code +1★ 299.2k
Matt Pocock's engineering skills thumbnailSKILL

Matt Pocock's engineering skills

Real engineering skills from a well-known TypeScript educator's own .agents folder.

Claude Code logoCodex logoClaude Code +1★ 291.4k
Ponytail: make your agent write less code thumbnailSKILL

Ponytail: make your agent write less code

A skill that pushes coding agents to act like a lazy senior: fewer lines, more tests, lower cost — without skipping the hard checks.

Claude Code logoCodex logoCursor logoClaude Code +2★ 164.8k

Free newsletter

Friday Picks

New AI tools you can use right away, in plain words. Every Friday.

Free. No spam. Unsubscribe anytime.