azure-ai-contentsafety-ts
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS] (SAFE): Dependencies include official Microsoft Azure packages (@azure-rest/ai-content-safety, @azure/identity, @azure/core-auth) from a trusted organization.
- [CREDENTIALS_UNSAFE] (SAFE): The skill correctly handles authentication by using environment variables and suggesting Managed Identity through DefaultAzureCredential, explicitly warning against hardcoding keys.
- [DATA_EXFILTRATION] (SAFE): Data transmission is restricted to legitimate Azure AI endpoints (*.cognitiveservices.azure.com). File system access is limited to reading images for the purpose of moderation, which is the skill's primary function.
Audit Metadata