scaling-facebook-ads
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources which presents a potential attack surface for indirect prompt injection.
- Ingestion points: The skill fetches content from user-provided URLs (competitor sites, product pages) and processes pasted or uploaded CSV campaign data.
- Boundary markers: The skill contains explicit instructions to treat retrieved content as data, never as instructions. It references
references/agent-security.mdwhich mandates quoting injection attempts and halting before the influenced step. - Capability inventory: The skill has the ability to read and write local state files (in the
.agents/directory), perform network fetches for research, and execute platform-specific tools like the Meta Ad Library. - Sanitization: There are explicit rules to escape interpolated values in any emitted markup and to redact sensitive tokens if encountered in input data.
Audit Metadata