The Agent Skills Directory

[SAFE]: The skill is focused on code quality and performance auditing. Its logic and tools (Grep, Read, Glob, Bash) are appropriately scoped for local filesystem analysis.
[PROMPT_INJECTION]: The skill processes untrusted code files at runtime, which represents an indirect prompt injection surface. This is a common characteristic for code-review tools.
Ingestion points: Local source code files in directories like src/ as referenced in the examples.
Boundary markers: No specific delimiters or safety instructions were found to separate code content from instructions.
Capability inventory: Bash (used for grep and sort), Read, Grep, and Glob tools.
Sanitization: No sanitization logic was found for the content of the files being reviewed.

code-reviewer