Find the right skill for your coding agent.

How the safety rating reads

Every listing is scanned as text, and each badge says what was covered. It is a screen for known risky patterns, not a guarantee. What is and isn’t checked

No issues found

The scan read the listing text and found none of the patterns it checks for. The badge names which files it read.

Review suggested

Something worth a look, such as an install script or an unusual instruction. Each finding shows the line and why.

Flagged

A pattern strongly associated with malicious skills, or a repository GitHub has blocked. Read the findings before installing.