Launching today

ContextOS

Launching today

Intelligent context routing for AI coding assistants

1 follower

Intelligent context routing for AI coding assistants

1 follower

Visit website

AI coding assistants waste thousands of tokens searching large codebases, often missing the files that actually matter. ContextOS indexes your repository into a semantic graph so AI Agents retrieve only the relevant functions, classes, documentation, and dependencies—reducing token usage while improving accuracy. Instead of sending entire files, ContextOS sends only the code the model actually needs.

Free

Launch tags:Developer Tools•Artificial Intelligence•GitHub

Launch Team / Built With

Viktor.comAn AI coworker that actually does the work

Promoted

Maker

📌

ContextOS started from a simple frustration: AI coding agents are powerful, but they keep forgetting the context that actually matters. So I built a context engine for them. It gives tools like Claude Code and Cursor persistent, structured memory across files, sessions, and repositories—while trying to keep the context window focused instead of dumping the entire codebase into it. ContextOS is open source, and this is still early. I’d genuinely love feedback from people building with coding agents every day—especially on what breaks, what feels unnecessary, and what you’d want it to remember next. Thanks for checking it out! 🚀

Report

23h ago