The Agent Skills Directory

[SAFE]: The skill implements a logical security assessment workflow (discovery, verification, and reporting) without any hidden or malicious side effects.
[SAFE]: No data exfiltration or unauthorized network communication was detected. The skill only interacts with project-specific analysis metadata and files in the sast/ directory.
[SAFE]: Instructions provided to subagents are clear and task-oriented, focusing on vulnerability identification without attempting to bypass AI safety guidelines or system constraints.
[SAFE]: Example command-line tools like curl are provided in documentation blocks for manual user verification and are not part of an automated or dangerous execution chain.

sast-fileupload