The Agent Skills Directory

[SAFE]: No active security risks were detected. The skill operates as a reference library using standard file-reading tools (Read, Grep, Glob) for its intended analysis workflow.
[COMMAND_EXECUTION]: The skill documents dangerous command execution patterns (e.g., in PHP and Ruby) to provide the agent with a reference for security auditing. These snippets are provided as static educational examples within markdown documentation.
[REMOTE_CODE_EXECUTION]: The skill includes reference material for unsafe functions (e.g., eval, pickle.loads) to serve as templates for vulnerability detection in external codebases.
[PROMPT_INJECTION]: While the skill is designed to ingest and analyze external files, creating a theoretical indirect prompt injection surface, it lacks the capabilities (such as network access or file-writing) required to facilitate an exploit.

sharp-edges