trademark-similarity-judgment
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data, including uploaded images and documents (Word, PDF, Excel), which serves as the primary ingestion point for analysis. * Ingestion points: Document and image uploads specified in SKILL.md. * Boundary markers: There are no explicit instructions to the AI to ignore instructions embedded within the user-provided data. * Capability inventory: Capabilities are restricted to trademark comparisons and data retrieval via MCP tools; no dangerous system commands or file-system writing capabilities were found. * Sanitization: No explicit input sanitization or filtering logic is provided.
- [SAFE]: All external references, connectivity checks, and documentation links point exclusively to the official domains of the author (open.zhihuiya.com and open.patsnap.com). These resources are essential for the skill's trademark search and analysis functions.
Audit Metadata