Open source
Ouroboros: Interview Your AI Before It Codes
Ouroboros is an open source Agent OS that interviews you before your AI coding agent writes code. It turns a vague request into a clear spec, then runs and checks the work.
Watch on YouTube · Watch on TikTok
The story
What the video said
This open source GitHub project turns a vague idea into a clear plan your AI coding agent can follow. You start with a rough request, and it asks you questions first.
You want your AI coding tool to build the right thing, but it often guesses wrong and wastes your time. You’ve tried writing longer prompts, but the agent still misses what you meant.
Ouroboros is an open source project that runs a short interview before your agent starts. It asks about your goal and scope, then writes a spec file you can run.
It scores how unclear your request is, and only builds the spec once that score drops low. Then it sends the spec to your agent and checks the result.
In the demo, you see the interview questions, a generated spec for a racing game, and a status panel listing each acceptance check as done, running, or waiting.
This page was generated from the Ming Dao AI publication record.
Ming Dao AI is independent and unaffiliated with GitHub or the featured project. Read our editorial policy.