meta-ad-policy-checker
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches live advertising policy content from Meta's official transparency domain (transparency.meta.com) to ensure analysis matches current standards.\n- [PROMPT_INJECTION]: The skill processes untrusted ad copy and external landing page content, which creates a surface for indirect prompt injection. 1. Ingestion points: Ad copy and advertiser context (Phase 0), landing page content (Phase 4). 2. Boundary markers: No explicit delimiters separate external data from agent logic. 3. Capability inventory: Performs network reads (fetching policies and landing page content) and generates text (analysis and rewrites). 4. Sanitization: No explicit sanitization or filtering of external content is mentioned.
Audit Metadata