The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands including git rev-parse, git log, and git diff to identify and retrieve code changes for analysis. These operations are essential for the skill's purpose.
[PROMPT_INJECTION]: The reviewer template in code-reviewer.md processes untrusted data in the form of code diffs, creating an indirect prompt injection surface.
Ingestion points: Code content and history retrieved via git diff and git log commands.
Boundary markers: The template uses markdown headers to separate sections, but it lacks specific delimiters or explicit instructions to ignore potentially malicious instructions embedded within the code diff text.
Capability inventory: The agent can execute git commands and dispatch subagents.
Sanitization: No sanitization or filtering of the code diff content is performed before the reviewer agent processes the data.

requesting-code-review