The Agent Skills Directory

[SAFE]: The skill is designed for legitimate search engine optimization tasks. All scripts and instructions are well-documented and consistent with the stated purpose of improving site visibility in Google Search and AI-driven surfaces.
[COMMAND_EXECUTION]: The scripts/test_skill.py script uses subprocess.run to execute local components, specifically scripts/validate.py and scripts/audit_page.py. This is a standard testing pattern used to verify the skill's own logic and environment and does not involve arbitrary command execution.
[EXTERNAL_DOWNLOADS]: The scripts/audit_page.py script utilizes the standard urllib.request library to fetch HTML content from user-provided URLs. This functionality is essential for its core purpose of auditing live web pages for SEO signals and does not involve downloading executable scripts or packages.
[INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface via scripts/audit_page.py, which reads and parses external HTML. The script uses the built-in html.parser.HTMLParser and json libraries to extract metadata and JSON-LD structured data. This data is converted into a structured JSON report, which is a safe way to handle untrusted external content without direct prompt interpolation or execution.

google-search-ai-optimization