search-health
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to audit external websites and verify guidance from official search-engine sources. This creates a surface where malicious instructions hidden in the audited site's source code (e.g., HTML comments, metadata) or compromised documentation could attempt to influence the agent's logic.
- Ingestion points: Target websites being audited, representative rendered pages, and official search-engine documentation (references to "official sources").
- Boundary markers: No specific boundary markers or "ignore instructions" delimiters are defined for the data being ingested.
- Capability inventory: The skill produces text-based deliverables (executive diagnosis, migration maps, prioritized backlogs). It does not include scripts for automated file writing, shell execution, or unauthorized network operations, although it mentions providing "verification commands or tools" for the user.
- Sanitization: The instructions do not prescribe specific sanitization or filtering steps for the content fetched from external sources before it is analyzed by the agent.
Audit Metadata