ohad6k/ditto

README preview

<p align="center"><img src="assets/ditto.png" width="360" alt="Ditto"></p>

<h1 align="center">Ditto</h1>

<p align="center"><b>Your AI agents act like they just met you. Ditto fixes that.</b></p>

<p align="center">
<img src="https://img.shields.io/github/stars/ohad6k/ditto?style=for-the-badge&color=3a3a3a&labelColor=141414&logo=github&logoColor=white&cacheSeconds=1800" alt="stars">
<a href="https://discord.gg/VNnMq2U5r"><img src="https://img.shields.io/badge/discord-join-3a3a3a?style=for-the-badge&labelColor=141414&logo=discord&logoColor=white" alt="discord"></a>
<img src="https://img.shields.io/badge/license-MIT-3a3a3a?style=for-the-badge&labelColor=141414" alt="MIT">
<img src="https://img.shields.io/badge/python-zero_deps-3a3a3a?style=for-the-badge&labelColor=141414&logo=python&logoColor=white" alt="Python, zero dependencies">
<img src="https://img.shields.io/badge/works_with-claude_·_codex_·_cursor_·_copilot-141414?style=for-the-badge&labelColor=3a3a3a" alt="works with claude, codex, cursor, copilot">
</p>

Your real coding-agent sessions already contain the rules you never wrote down: what “done” means, what you reject on sight, how you debug, how you design UI, and how you write when you are actually working.

Ditto mines selected evidence from those sessions — Claude Code, Codex, and Copilot CLI logs out of the box — into a private working profile your agent reads before every task. Separate layers for work, design, and writing mean the right part of you loads for the right task.

## Not memory

View full repository on GitHub →