The Agent Skills Directory

[SAFE]: The skill consists of high-level guidance for pull request reviews and security audits. No scripts, command-line executions, or network requests are defined within the skill content.
[SAFE]: No obfuscation, prompt injection attempts, or metadata poisoning were detected. The skill instructions are consistent with its stated purpose of improving code quality.
[SAFE]: While the skill is intended to process external code (untrusted data), it possesses no tool capabilities or scripts, effectively eliminating the risk of indirect prompt injection being leveraged for malicious actions within the skill's own scope.

code-reviewer