The Agent Skills Directory

[COMMAND_EXECUTION]: The skill incorporates a suite of regular expression patterns designed to identify dangerous system operations, such as recursive deletions and arbitrary code execution via eval or exec. These patterns serve as the signature database for the scanner and are not executed by the skill itself during the auditing process.
[DATA_EXFILTRATION]: The scanner contains detection logic for identifying data exfiltration attempts, including monitoring for access to sensitive credential paths like .ssh and .aws directories, as well as unauthorized environment variable scraping. These markers are static data used for analysis and do not represent functional vulnerabilities within the skill.
[EXTERNAL_DOWNLOADS]: The skill references the Streamlit library for its optional graphical user interface. Streamlit is a widely recognized and well-known service for Python application development, and its inclusion follows standard practices for creating interactive tools.

skill-scanner