The Agent Skills Directory

[SAFE]: The skill defines formatting rules for code reviews and does not include scripts, remote downloads, or sensitive data access patterns.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it is designed to process external code diffs.
Ingestion points: Processes pull request diffs and descriptions as indicated in the SKILL.md instructions.
Boundary markers: Absent; the skill does not define specific delimiters to isolate untrusted code diffs from agent instructions.
Capability inventory: No high-risk capabilities (subprocesses, file writes, or network access) are present; the skill is limited to generating text output.
Sanitization: No sanitization or validation of the input code is specified.

caveman-review