meta-ads
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could theoretically be used to influence the agent's behavior during content generation tasks.
- Ingestion points: Lead data captured via website forms or Meta Instant Forms, and customer comments on Facebook/Instagram ads (Section 10).
- Boundary markers: The instructions include clear prohibitions against sending personal data in custom tracking parameters (Section 3).
- Capability inventory: The agent has the ability to draft replies to comments and generate new ad creative based on historical lead performance.
- Sanitization: The skill mandates SHA-256 hashing for sensitive identifiers like email and phone numbers before they leave the server environment.
- [SAFE]: Network operations are restricted to the official Meta Graph API (graph.facebook.com), which is a well-known and expected service for this skill's functionality.
- [SAFE]: The skill enforces a 'Human-in-the-Loop' model, explicitly requiring written owner approval for any action involving financial expenditure or public posting.
Audit Metadata