google-ads-transparency-center
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local bash script to process user inputs and communicate with the Frevana API.
- [EXTERNAL_DOWNLOADS]: The skill performs network requests to
ai-factory.frevana.comto retrieve ad creative information. - [INDIRECT_PROMPT_INJECTION]: The skill ingests ad creative content (text, advertiser names, etc.) from the Google Ads Transparency Center, creating an attack surface for instructions embedded in external data.
- Ingestion points: Data enters the system via the Frevana API response in
scripts/search_google_ads_transparency_center.sh. - Boundary markers: The instructions do not define specific delimiters for separating untrusted ad content from agent instructions during summarization.
- Capability inventory: The skill has the ability to write JSON results to the local filesystem and perform network operations to the vendor API.
- Sanitization: The script includes JSON structure validation using Python to ensure the response is well-formed before it is returned to the agent.
Audit Metadata