The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses standard terminal utilities such as curl, jq, grep, and sed to analyze website headers, robots.txt, and sitemaps. These are used strictly for read-only analysis of public web resources.
[EXTERNAL_DOWNLOADS]: The skill fetches configuration and asset data from the target website being audited (e.g., robots.txt, sitemap.xml) and uses npx to run the well-known lighthouse auditing tool from the official NPM registry. These operations are essential to the skill's primary purpose and target well-known, trusted services.
[DYNAMIC_EXECUTION]: The skill includes a Python script (seo_page_check.py) for parsing HTML signals. This script uses the standard library's html.parser and urllib to fetch and analyze page metadata. It does not execute arbitrary code or use unsafe functions like eval().
[SAFE]: All network operations are directed at the user-specified SITE or well-known APIs (Google CrUX). The instructions prioritize user privacy by advising against signing into services and instead requesting manual exports for sensitive data.

seo-audit