The Agent Skills Directory

[SAFE]: The skill instructions are focused on legitimate SEO analysis tasks. It directs users to vendor-provided documentation on GitHub (github.com/PapiScholz/search-visibility-kit) for configuring the required MCP tools, ensuring that sensitive access is managed through standard client configurations rather than the skill instructions themselves.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources, which is a common requirement for SEO tools but presents an attack surface for indirect prompt injection.
Ingestion points: External sitemap files are processed in SKILL.md (via list_sitemaps_enhanced) to compare URL counts, and search performance metrics containing user queries are retrieved from Google Search Console.
Boundary markers: None present; the instructions do not specify delimiters to separate external data from agent instructions.
Capability inventory: The skill utilizes tools like submit_sitemap and run_report which interact with external Google service APIs.
Sanitization: There are no explicit instructions for sanitizing or escaping content retrieved from target websites.

seo-audit