Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill processes content from external PDF files, which represents a potential attack surface for indirect prompt injection. Malicious instructions embedded in a PDF could be extracted and processed by the agent. Ingestion points: PDF text extraction using pypdf, pdfplumber, and pytesseract. Boundary markers: None present in the provided examples. Capability inventory: Local file operations and system command execution for PDF tools. Sanitization: No content sanitization is demonstrated in the examples.
- [EXTERNAL_DOWNLOADS]: The documentation references legitimate and widely-used Python packages and system utilities. Evidence: Mentions pypdf, pdfplumber, pandas, reportlab, pytesseract, and pdf2image, as well as poppler-utils, qpdf, and pdftk.
Audit Metadata