The Agent Skills Directory

Indirect Prompt Injection Surface: The skill is designed to process external code changes and repository diffs. This creates a potential surface where instructions embedded in the code being reviewed could attempt to influence the agent's behavior.
Ingestion points: The skill reads pull request diffs and repository files as specified in the implementation rules (SKILL.md).
Boundary markers: No specific delimiters or instructions to ignore embedded commands are defined in the provided text.
Capability inventory: The skill uses repository file-reading, search tools, and git commands, but it does not specify high-risk capabilities such as external network requests or file system writing.
Sanitization: The instructions do not include specific methods for sanitizing or escaping the content being analyzed.

code-review