The Agent Skills Directory

[SAFE]: The skill is a defensive tool focused on security auditing. It provides structured instructions for detecting various security flaws using standard filesystem and search tools.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted source code for analysis. However, this behavior is essential to its primary function as a security scanner and does not indicate malicious intent.
Ingestion points: Files are accessed via Read, Grep, and Glob tools.
Boundary markers: None explicitly defined in the provided file.
Capability inventory: The skill uses Bash, Read, Grep, and Glob for file analysis.
Sanitization: Content is analyzed via pattern matching; no specific sanitization of input data is mentioned before processing.

rc-codeprobe-security