pdf-reading

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for the agent to use various command-line utilities (such as pdfinfo, pdftotext, pdfimages, pdfdetach, pdffonts, pdftoppm, and pdftk) to analyze and process PDF documents.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted PDF files which may contain embedded instructions designed to influence agent behavior.
  • Ingestion points: Data is read from external PDF files using tools like PdfReader and pdftotext.
  • Boundary markers: The provided code snippets do not include explicit delimiters or instructions to the agent to ignore text that looks like a prompt inside the PDF.
  • Capability inventory: The skill has the capability to run shell commands and write files to the /tmp directory.
  • Sanitization: The skill documentation includes a specific recommendation to use os.path.basename() to sanitize filenames extracted from PDF attachments, which mitigates path traversal risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 04:17 AM
Security Audit — agent-trust-hub — pdf-reading