The Agent Skills Directory

[SAFE]: The skill is designed for the singular purpose of performing SEO audits. It follows a structured 12-point checklist defined in the documentation and operates on user-provided inputs without bypassing security controls.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface as it ingests untrusted content from external web pages.
Ingestion points: The skill fetches the HTML content of external URLs using the WebFetch tool.
Boundary markers: There are no explicit delimiters or instructions to the agent to disregard potential instructions embedded within the fetched HTML.
Capability inventory: The skill has the capability to write files to the local directory (generating an HTML report) and access the seo-web-cafe MCP tool.
Sanitization: No explicit sanitization or filtering of the fetched HTML is described before the content is analyzed or placed in the report template.
[COMMAND_EXECUTION]: The skill generates and saves a local HTML file (seo-audit-[domain]-[date].html) to the current working directory. This file is constructed using a static template and data extracted during the audit; it does not involve the execution of arbitrary shell commands or malicious scripts.

page-seo-checker