What it does
Composes Unix-style tools that seed, expand, judge, verify, and report code defects while preserving evidence chains.
Agent tooling
Only you can see your notes.
Screenshot unavailable. Open the experiment ↗
Composes Unix-style tools that seed, expand, judge, verify, and report code defects while preserving evidence chains.
A developer can run these tools to hunt for hidden bugs in your project. The software scans new code changes to find risky patterns and gathers related files. It then asks Jev, an outside AI service, to rate the risk level.
Your developer needs an access key generated by an external AI service to connect the tools. Before building a final report, the software checks the AI claims against your actual files. The tools drop the warning if the AI invents a file that does not exist.