The Agent Skills Directory

[SAFE]: The skill is designed to perform static analysis of local HTML files. It does not perform network operations, access sensitive credentials, or attempt to persist on the system.
[COMMAND_EXECUTION]: The scripts/analyze.sh script uses standard shell commands (grep, find, jq) to inspect HTML content. It implements security best practices, such as using set -euo pipefail for robust error handling and null-delimited processing (-print0) to safely handle files with special characters or spaces in their names.
[EXTERNAL_DOWNLOADS]: The skill does not download external scripts or packages. It relies on the pre-installed system utility jq to generate structured JSON output.
[DATA_EXFILTRATION]: There are no indicators of data exfiltration. The script reads local files but only outputs the audit findings to the standard output (stdout) for the agent's use.
[PROMPT_INJECTION]: The SKILL.md file contains instructional content and audit criteria that are consistent with its stated purpose. No malicious override patterns or system prompt extraction attempts were detected.

web-quality-audit