google-ai-overview
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE_EXFILTRATION]: The skill transmits a page token to a vendor-controlled domain (ai-factory.frevana.com). This is a functional requirement for the skill and does not constitute unauthorized exfiltration.
- [CREDENTIALS_UNSAFE]: The skill requires an API token for authentication. It correctly manages this secret by looking for it in environment variables or requesting it via secure user input, avoiding unsafe hardcoding.
- [REMOTE_CODE_EXECUTION]: All scripting logic is provided as local, hardcoded Bash and Python code. The skill does not download or execute scripts from remote or untrusted sources.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from an external API response. It includes explicit instructions for the agent to return the response unchanged, which serves as a boundary marker to prevent the agent from being influenced by potentially malicious content within the API output.
Audit Metadata