The Agent Skills Directory

[SAFE]: The skill operates as a static analyzer for test suites. It utilizes the Read, Grep, Glob, and Bash tools exclusively for standard development workflows, such as checking git status and reading source files.
[INDIRECT_PROMPT_INJECTION]: The skill represents an indirect prompt injection surface because it processes untrusted source code and test files. However, it specifically notes that it performs static analysis only and does not execute the tests, which mitigates risks of remote code execution during the review process.
Ingestion points: Reads code and test files via the Read tool.
Boundary markers: None explicitly defined in instructions.
Capability inventory: Uses Bash (restricted to git commands in instructions), Read, Grep, and Glob tools.
Sanitization: Not explicitly mentioned, but behavior is restricted to generating a static review report.

test-reviewer