research
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes information from external primary sources which could contain malicious content. 1. Ingestion points: Data is fetched from primary sources and APIs in SKILL.md. 2. Boundary markers: A Judge agent in SKILL.md Step 3 provides validation. 3. Capability inventory: The skill is authorized to write Markdown reports to the filesystem in SKILL.md Step 5. 4. Sanitization: No technical sanitization is mentioned.
- [SAFE]: The skill is a set of instructions for agent orchestration with clear safety boundaries, such as the prohibition against modifying product code.
Audit Metadata