Skip to main content

Module security

Module security 

Source
Expand description

Leak detection for agent output before it is posted to GitHub. It uses the strict rules of secret-redact, so a low-confidence pattern blocks a post as much as a known token does

Functionsยง

has_leaks
Whether text holds anything that looks like a secret. Nothing is posted to GitHub when it does
redact
text with every secret replaced by [REDACTED:{rule}]
scan
Every leaked secret in text, with its rule and line