ads-intelligence
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes advertising copy and creative text from external ad networks, which constitutes an ingestion surface for potential indirect prompt injection attacks.\n
- Ingestion points: Ad content and creative text retrieved from external search endpoints via api.leadmagic.io (SKILL.md).\n
- Boundary markers: No specific delimiters or instructions are provided to the agent to treat returned ad copy as untrusted data.\n
- Capability inventory: Performs authenticated network requests using API keys provided through environment variables (SKILL.md).\n
- Sanitization: Ad copy is retrieved and processed without explicit sanitization or filtering for embedded instructions.
Audit Metadata