The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data in the form of search queries from Google Search Console. While this presents a theoretical surface for indirect prompt injection, the risk is mitigated by workflow requirements.
Ingestion points: Search query data processed from get_branded_performance and query_top_queries tools.
Boundary markers: The skill does not explicitly define delimiters for external data, but uses strict interpretation and grounding rules.
Capability inventory: Includes update_site which modifies site configuration, alongside several read-only diagnostic tools.
Sanitization: No explicit sanitization of query strings is described, but the agent is instructed to manually scan and filter variants before proposing them for a write operation.

branded-split