How to use COUTCHECK: scan output for leaks before you send it

COUTCHECK scans a response for PII and toxicity, optionally validating JSON, and returns the entities found so you can redact, block, or regenerate.

TarkovaJun 5, 20263 min read

The response is the other trust boundary. COUTCHECK catches a leaked email or a toxic line before your user sees it.

crowkis cli

> COUTCHECK "Sure! Email john.doe@acme.com for the refund."
1) "fail"
2) "pii: [email:john.doe@acme.com]"
3) "toxic: []"
4) "json: n/a"

Act on the verdict per failure mode, redact the PII and continue, or regenerate.

Keep reading