aj-patent-disclosure-cn
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [SAFE]: The skill's instructions, templates, and README are strictly focused on the patent disclosure workflow. There are no attempts to override safety filters, extract system prompts, or exfiltrate sensitive user data.
- [COMMAND_EXECUTION]: Python scripts such as
scripts/create_cnipa_reference_doc.pyandscripts/generate_figures.pyusesubprocess.runto call system utilities likepandocand the Mermaid CLI (mmdc). These calls are used for legitimate document formatting and diagram rendering tasks and do not pass unsanitized user input to the shell. - [EXTERNAL_DOWNLOADS]: The
scripts/setup_env.shutility script manages the installation of well-known Python packages (pypandoc,matplotlib) from a reputable repository mirror. It does not perform any unverified remote script execution (e.g., pipe to bash). - [SAFE]: The
scripts/office_to_markdown.pyscript is designed to extract text from Office files (.docx,.pptx). It explicitly implements security measures by not executing document macros, external links, or embedded objects, and utilizes safe path joining techniques to prevent directory traversal during media extraction.
Audit Metadata