The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources (Google Search Console and Bing Webmaster Tools) which could potentially contain malicious content intended to influence the agent's behavior.
Ingestion points: Untrusted data enters the agent context through the outputs of list_sitemaps, get_sitemap_performance, list_url_inspections, and get_bing_feeds in SKILL.md.
Boundary markers: There are no explicit instructions to use delimiters or ignore embedded instructions within the ingested tool data.
Capability inventory: The skill possesses the capability to perform write operations (submitting sitemaps) via the submit_sitemap tool.
Sanitization: No explicit sanitization or validation of the data returned by the tools is described beyond mapping it into groups for report generation.
Mitigation: The risk is significantly mitigated by the requirement in Step 6 to list exact URLs and ask the user for confirmation before calling the submit_sitemap tool.

sitemap-audit