azure-ai-contentsafety-java
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill incorporates the official Azure AI Content Safety SDK (com.azure:azure-ai-contentsafety). This dependency is sourced from a well-known service and organization, presenting no security risk.
- [CREDENTIALS_UNSAFE]: The skill follows security best practices by recommending the retrieval of sensitive API keys and endpoints from environment variables (CONTENT_SAFETY_KEY, CONTENT_SAFETY_ENDPOINT) instead of hardcoding them in the source code.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external text and images for moderation purposes. Ingestion points include text analysis options, local image files, and remote image URLs. While boundary markers and sanitization are absent, the primary capability is sending data to the Azure API for analysis. Given the primary purpose is content safety, the associated risk is minimal.
Audit Metadata