chinese-ppt2
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns detected. The skill follows best practices for LaTeX document generation and provides templates for academic use cases.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by design, as it processes user-provided research papers (PDF/docx) to extract data and images. 1. Ingestion points: User-uploaded research papers (PDF/docx) used for content extraction (SKILL.md, '工作流'). 2. Boundary markers: Absent; no instructions are provided to ignore embedded commands within the source documents. 3. Capability inventory: Local file system writes (.tex and /figures/) and automated parsing logic via PyMuPDF. 4. Sanitization: Absent; extracted text is interpolated into LaTeX templates using regex and library calls.
Audit Metadata