The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute standard development tools including git diff, git show, git status, and git log, as well as cat for reading local project guidelines. These operations are scoped to the current repository and are necessary for the skill's stated purpose of reviewing code changes.
[PROMPT_INJECTION]: The skill processes untrusted data from the repository's git history and source files (indirect prompt injection surface). However, this risk is effectively mitigated by mandatory human-in-the-loop checkpoints: Step 4 and Step 5A explicitly require user approval before the agent can apply any code changes or fixes.
[DATA_EXPOSURE]: The skill reads project-specific guideline files (CLAUDE.md, AGENT.md) to provide context-aware reviews. This data is handled within the local session and is not transmitted to external or untrusted domains.

code-review