industry-competition-moat
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to ingest and analyze data from external sources including public policy documents, statistics, and company disclosures. While this creates a surface for indirect prompt injection, the skill does not possess dangerous capabilities such as arbitrary command execution, network exfiltration of local data, or file writing.
- Ingestion points: Public policy websites (MIIT, NDRC), statistical bureaus, procurement portals, and exchange disclosures.
- Boundary markers: The skill instructs the agent to "Separate company claims from third-party validation" and "State which competitor comparison points are fact and which are inference," providing a structural framework for validation but no technical delimiters.
- Capability inventory: No subprocess calls, network POST operations, or file system modifications are present in the skill files.
- Sanitization: No explicit sanitization logic for external content is provided.
Audit Metadata