mistral-data-handling

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides utility code for redacting Personally Identifiable Information (PII) such as emails, phone numbers, and SSNs from text before sending it to third-party APIs.
  • [SAFE]: The skill follows security best practices by recommending the use of environment variables (process.env.MISTRAL_API_KEY) for managing API credentials rather than hardcoding them.
  • [SAFE]: File system operations (fs.createReadStream, fs.createWriteStream) are used appropriately for the stated purpose of sanitizing local training datasets.
  • [SAFE]: All external references point to official Mistral AI documentation and privacy policies.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 02:27 PM