nature-paper-to-patent

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious code, obfuscation, or data exfiltration attempts were detected across the skill's scripts or instructions. The workflow is grounded in technical evidence and uses systematic validation gates.- [COMMAND_EXECUTION]: The script scripts/build_patent_package.py utilizes subprocess.run to orchestrate internal tasks by calling other local Python scripts within the skill folder using sys.executable. This is a legitimate implementation for a multi-stage toolchain and is executed without shell involvement.- [EXTERNAL_DOWNLOADS]: Dependencies listed in requirements.txt (latex2mathml, Pillow, pypdf, and python-docx) are well-known, standard libraries used for PDF extraction, image rendering, and document generation.- [SAFE]: The skill employs dynamic loading of local modules using importlib.util in scripts/build_patent_package.py and scripts/render_patent_docx.py. This is used purely for modular code organization and is restricted to files within the skill's local directory structure.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 03:03 PM
Security Audit — agent-trust-hub — nature-paper-to-patent