markitdown

Pass

Audited by Gen Agent Trust Hub on Jun 14, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Downloads the 'markitdown' library from Microsoft's official GitHub repository and utilizes well-known services like OpenRouter for AI-enhanced document processing.
  • [COMMAND_EXECUTION]: Uses 'subprocess.run' in 'scripts/generate_schematic.py' to call a local helper script. The command is constructed as a list of arguments, which prevents shell injection vulnerabilities.
  • [SAFE]: The skill identifies a potential surface for indirect prompt injection as it processes content from untrusted documents (PDFs, Word files, etc.) and passes it to LLMs. This is a standard risk for document processing tools and is mitigated by the safety guardrails of the underlying language models.
  • [SAFE]: Interacts with reputable AI models via the OpenRouter API. API keys are managed through environment variables or local configuration, following standard security practices.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 14, 2026, 09:23 AM
Security Audit — agent-trust-hub — markitdown