The Agent Skills Directory

[Data Exposure] (SAFE): The accessibility_checker.py script reads local HTML/JSX/TSX files within a specified project directory to perform accessibility audits. It does not access sensitive system files (e.g., SSH keys, env files) and contains no network exfiltration logic.
[Remote Code Execution] (SAFE): No patterns of remote script downloading (curl/wget) or execution (bash/python pipe) were found. The script uses only Python standard libraries.
[Command Execution] (SAFE): The skill does not utilize dangerous functions such as eval(), exec(), os.system(), or subprocess. It operates solely as a static analysis tool for code files.
[Obfuscation] (SAFE): No hidden, encoded, or obfuscated content (e.g., Base64, zero-width characters) was detected in the scripts or markdown documentation.
[Indirect Prompt Injection] (LOW): While the script processes untrusted project files, it performs static regex-based analysis and does not interpret file content as instructions for the agent, nor does it have capabilities that could be exploited through the data it reads.

frontend-design