alterlab-pdf-extract

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of markitdown (from Microsoft) and the openai Python client. These are standard, well-known libraries for document processing and AI interaction.
  • [PROMPT_INJECTION]: The extract_llm function in scripts/extract_to_table.py interpolates document content directly into a prompt sent to an LLM. This presents a surface for indirect prompt injection if a processed document contains malicious instructions, though this is inherent to the tool's core functionality.
  • Ingestion points: Document text is read and converted via markitdown in the extract_llm function within scripts/extract_to_table.py.
  • Boundary markers: The prompt uses basic labels like Paper ({source}): to delineate content, which provides minimal protection against adversarial content.
  • Capability inventory: The skill possesses the capability to read local documents, write output tables, and perform network requests to LLM providers (OpenRouter/OpenAI).
  • Sanitization: No specific sanitization or escaping of the document content is performed before interpolation into the prompt.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 03:28 PM
Security Audit — agent-trust-hub — alterlab-pdf-extract