The Agent Skills Directory

[SAFE]: The skill operates as a code auditor, utilizing local scripts and standard command-line tools like git and the GitHub CLI (gh) to inspect the repository. These operations are consistent with its stated purpose of performing PR reviews and skill validation.
[SAFE]: It implements proactive security measures by auditing other skills for unsafe credential handling, specifically flagging any attempts to directly read .env files as blocking violations.
[SAFE]: The included Python scripts for validating evaluation schemas and trigger overlaps utilize safe path handling practices, such as absolute path resolution and relative check constraints, to mitigate directory traversal risks.
[SAFE]: External tool integration (the skill-validator binary) is handled via a wrapper script that does not perform automatic downloads, instead providing installation instructions for the user. Network operations are limited to standard GitHub interactions via gh.

confluent-skill-reviewer