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
Risk Level
SAFE
Analyzed
Sep 16, 2026, 09:16 AM
Security Audit — agent-trust-hub — industry-competition-moat