skills/nexscope-ai/nexscope-ecommerce-skills/ecommerce-ruiguan-detection-patent-design/Gen Agent Trust Hub
ecommerce-ruiguan-detection-patent-design
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill includes Python scripts (
ruiguan_detection_patent_design.pyandupload_image.py) designed to be executed by the agent. These scripts facilitate communication with the vendor's patent detection and asset management APIs using Python's standardurlliblibrary. - [EXTERNAL_DOWNLOADS]: The scripts perform network operations to communicate with Nexscope's API infrastructure (configured via the
NEXSCOPE_PROXY_BASEenvironment variable). This includes sending product images for analysis and submitting feedback to the vendor's feedback endpoint. These are official vendor resources required for the skill's primary functionality. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data, such as
imageUrl,productTitle, andproductDescription. This data is passed as parameters to the external patent detection API. The vulnerability surface is limited as the scripts handle the data as structured JSON for API consumption rather than directly interpolating it into a downstream LLM prompt. Evidence:imageUrl,productTitle, andproductDescriptionparameters inSKILL.mdandscripts/ruiguan_detection_patent_design.py.
Audit Metadata