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