The Agent Skills Directory

[SAFE]: No security issues detected. The skill instructions and templates follow best practices for local code review workflows.
[COMMAND_EXECUTION]: The skill utilizes standard git commands (git rev-parse, git log, git diff) to facilitate the code review process. These commands are used locally to identify and compare commit history.
[PROMPT_INJECTION]: The skill is designed to ingest and review code diffs, which creates a surface for indirect prompt injection if the code being reviewed contains malicious instructions.
Ingestion points: git diff output is ingested into the code-reviewer.md template.
Boundary markers: None present; the code diff is directly inserted into the prompt template.
Capability inventory: Shell execution via git is the primary capability.
Sanitization: None; the skill assumes the code is safe to read for review purposes.

requesting-code-review