Confirmed-only AI review before you push
How it works
Why it beats the alternative
How it actually works
Exactly what it does — and when.
A developer points the CLI at the current Git diff and an authorized endpoint, enables the security lens, and inspects the resulting findings. They reproduce any reported issue and run project tests before using the exit status as part of their pre-push workflow.
Use it to…
Review before pushing
Point the CLI at the current Git diff and inspect confirmed findings before publication.
Add a security lens
Use the security option for a focused pass, then verify every reported issue against the code.
Gate a local workflow
Use the documented exit codes in a pre-push or CI step after testing the command in your environment.
The shift it creates
- $80–150/mo SaaS bills stacking up
- Rate caps + vendor lock-in
- Data you don't actually own
- Runnable single-file Node CLI, zero deps
- Reviews a git diff → confirmed findings only
- --security lens (vuln taxonomy + exclusions)
What's inside
Who it's for
Developers using Git locally
want a repeatable review command before pushing
Teams with OpenAI-compatible endpoints
can connect their own authorized model access
CI and hook maintainers
need documented exit codes after testing the workflow
Questions, answered
Ready to put AI Code Review — Runnable CLI to work?
Free download · no account needed · yours forever