The Agent Skills Directory

[SAFE]: The skill is designed to analyze code diffs and pull requests. While it processes untrusted external data, the skill lacks exploitable capabilities such as file system writes, network exfiltration tools, or dynamic code execution modules that could be abused through indirect prompt injection.
[SAFE]: No instances of obfuscation, credential exposure, or unauthorized command execution were found. The skill's instructions focus on providing high-quality, actionable feedback using established software engineering best practices.
[SAFE]: All external references in the documentation point to legitimate development workflows and project-specific guidelines from a trusted author context.

review-changes