paid-media-audit
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from untrusted external sources, including ad platform CSV exports and content retrieved through web searches (Meta Ad Library, industry blogs, and marketing forums like Reddit), creating a surface for indirect prompt injection.
- Ingestion points: User-provided ad data exports and retrieved content from the web and ad libraries.
- Boundary markers: The skill utilizes a dedicated security reference (
references/agent-security.md) that instructs the agent to treat all fetched content as data, never to follow instructions found within that data, and to quote suspected injection attempts. - Capability inventory: The skill has access to the
meta-ad-librarytool, web search capabilities, and the ability to read and write context files in the.agents/directory. - Sanitization: Explicit guidelines require HTML-escaping of interpolated values and validation of URLs in any emitted markup to prevent cross-site scripting (XSS).
- [EXTERNAL_DOWNLOADS]: The skill fetches competitor advertising insights and industry performance benchmarks from well-known platforms including the Meta Ad Library and marketing analytics providers.
Audit Metadata