mistral-ocr-markdown

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security vulnerabilities were detected in the skill code or instructions.
  • [DATA_EXFILTRATION]: The skill uploads document content to Mistral AI's official API (api.mistral.ai) for OCR processing. This behavior is documented, expected, and necessary for the skill's primary function.
  • [EXTERNAL_DOWNLOADS]: Dependencies are restricted to legitimate, well-known packages (@mistralai/mistralai and dotenv) sourced from the official registry.
  • [CREDENTIALS_UNSAFE]: The skill follows security best practices by resolving API keys through environment variables, CLI flags, or .env files, and does not contain hardcoded secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 11:52 PM
Security Audit — agent-trust-hub — mistral-ocr-markdown