marketing-ideas
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, exfiltration attempts, or harmful commands were found in the skill. The logic is purely instructional for providing marketing advice.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest context from local files like '.agents/product-marketing-context.md'. Evidence: 1. Ingestion points: Loads context from external project files. 2. Boundary markers: Absent. 3. Capability inventory: The skill is restricted to generating text-based marketing advice; no file system write or network capabilities are defined in this specific file. 4. Sanitization: Absent. This represents a minor ingestion surface for untrusted data, but the risk is negligible given the skill's scope.
Audit Metadata