The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses local git commands (rev-parse, diff, show, worktree) to manage and inspect code versions. These operations are standard for development workflows and are used here to provide context to a subagent.
[INDIRECT_PROMPT_INJECTION]: Because the subagent is instructed to read and analyze code changes, there is a theoretical risk of indirect prompt injection if the code being reviewed contains malicious instructions (e.g., in comments) designed to deceive the reviewer. This is an inherent risk for any tool that processes untrusted text via an LLM.

requesting-code-review