gemini-document-processing

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: Indirect prompt injection surface detected. The skill is designed to process external PDF documents and user-provided prompts using the Google Gemini API.
  • Ingestion points: The process-document.py script reads the content of local PDF files and user prompts provided via CLI arguments.
  • Boundary markers: The implementation does not utilize explicit boundary markers or instructions to ignore embedded commands within the processed PDF content.
  • Capability inventory: The skill performs network operations to the official Google Gemini API (generativelanguage.googleapis.com). It does not possess capabilities for arbitrary command execution, local file modification, or privilege escalation.
  • Sanitization: No specific filtering or sanitization is applied to the PDF content before it is transmitted to the model.
  • [EXTERNAL_DOWNLOADS]: The skill documentation and scripts specify dependencies on external Python packages including google-genai, python-dotenv, httpx, and pydantic. These are standard, well-known libraries for interacting with Google's AI services and managing application configuration. References to external PDF documents in the documentation target reputable domains such as nasa.gov and ucl.ac.uk.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 06:30 PM