The Agent Skills Directory

[SAFE]: The skill's primary functionality is implemented through local Python scripts (a11y_scanner.py and contrast_checker.py) that perform static analysis via regular expressions. The scripts are self-contained and do not invoke any external processes or dynamic code execution.\n- [EXTERNAL_DOWNLOADS]: The tools are designed to be 'stdlib-only', meaning they rely exclusively on the Python standard library. No external packages from PyPI or npm are required or downloaded during use.\n- [DATA_EXFILTRATION]: No network-related code (such as use of requests, urllib, or socket) was found in any of the scripts. The skill operates locally on the project files provided by the user.\n- [COMMAND_EXECUTION]: There are no instances of os.system, subprocess.run, or similar functions that would allow for arbitrary command execution. The scripts only read and parse text files.\n- [REMOTE_CODE_EXECUTION]: No patterns of remote script fetching or execution were detected. The skill uses local reference documentation and examples to guide remediation without connecting to remote servers.

a11y-audit