The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill operates on data retrieved from external sources (Google Search Console), which may contain untrusted content such as query strings, sitemap URLs, or site metadata.
Ingestion points: Search Analytics query results, sitemap contents, and site inspection data are ingested through the gog CLI tool commands described in SKILL.md.
Boundary markers: The skill proactively suggests the use of the --wrap-untrusted flag when processing Google content to help the agent distinguish between instructions and data.
Capability inventory: The skill allows the agent to query, list, and modify (submit/delete) Google Search Console data using the gog tool.
Sanitization: While the instructions advocate for safe flags like --wrap-untrusted and --readonly, the skill itself does not implement automated sanitization or schema validation for the external content it retrieves.

gog-searchconsole