duo-difficulty-calibration
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists solely of Markdown documentation files that explain theoretical concepts for joint item-user modeling and difficulty management. There are no executable scripts or system commands present in the provided files.
- [SAFE]: All external URLs are informational links to official Duolingo blog posts used as sources. No network exfiltration, credential harvesting, or communication with untrusted domains was detected.
- [SAFE]: The skill does not employ any obfuscation, dynamic execution, or persistence mechanisms. Metadata fields are used appropriately for versioning and author attribution without poisoning patterns.
Audit Metadata