document-processing

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill leverages established command-line tools such as pdftotext, tesseract, and pandoc to handle document conversion and character recognition. These operations are standard for the tool's primary purpose.
  • [EXTERNAL_DOWNLOADS]: The skill includes functionality to interact with the Mistral OCR API. This configuration involves transmitting document data to a well-known service provider's official API endpoint for processing.
  • [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes external, untrusted content from PDFs and images.
  • Ingestion points: Document files (PDF, PNG, DOCX) processed in SKILL.md and references/pdf-tools.md.
  • Boundary markers: None identified in the provided extraction examples.
  • Capability inventory: Shell command execution and network requests via httpx documented in SKILL.md.
  • Sanitization: No explicit validation or filtering of extracted text content is shown.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 05:54 AM
Security Audit — agent-trust-hub — document-processing