ai-paper-reader

Pass

Audited by Gen Agent Trust Hub on Oct 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of PDF research papers, which creates a surface for indirect prompt injection. • Ingestion points: The skill is designed to read paper.pdf files provided by the user. • Boundary markers: The prompt instructions do not include specific delimiters or warnings to the agent to ignore potentially malicious instructions embedded within the PDF text or metadata. • Capability inventory: The skill includes scripts that perform filesystem write operations (saving images to the images/ directory). • Sanitization: No sanitization or content validation is performed on the data extracted from the PDFs before it is interpreted by the agent.
  • [COMMAND_EXECUTION]: The skill provides functional Python scripts that utilize the PyMuPDF (fitz) library to perform local filesystem operations, such as creating directories and writing image files extracted from PDFs.
  • [EXTERNAL_DOWNLOADS]: The skill instructions recommend the installation of external Python packages, specifically pymupdf and pdf2image, which are standard, well-known libraries used for document processing tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 16, 2026, 03:03 AM