The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses git diff and git diff --name-only to retrieve code changes for analysis. These are standard read-only commands for code review tasks.
[SAFE]: The review rubric explicitly mandates checks for security risks like hardcoded secrets and injection vulnerabilities, reflecting a defensive design.
[PROMPT_INJECTION]: The skill processes untrusted code changes, establishing an indirect prompt injection surface. Ingestion point: git diff output. Boundary markers: absent. Capability inventory: git command execution. Sanitization: absent. This surface is inherent to the skill's function and no malicious intent was found.

deep-review