Moderates user-generated text and images for harmful content using the Azure AI Content Safety REST SDK.
This skill integrates the Azure AI Content Safety service into your TypeScript applications, enabling automated detection of hate speech, violence, sexual content, and self-harm. It provides a robust framework for building safer digital environments by offering granular severity levels for both text and image analysis. With support for custom blocklists and flexible authentication via API keys or Managed Identity, developers can implement sophisticated moderation workflows that align with specific community guidelines and compliance requirements.
主な機能
011,777 GitHub stars
02Image moderation via Base64 encoding or Blob URL references
03Customizable blocklist management for domain-specific prohibited terms
04Type-safe REST client implementation with built-in error handling guards
05Multi-category text analysis for hate, violence, and self-harm detection
06Configurable severity thresholds with 4-level or 8-level granularity
ユースケース
01Automating moderation for social media platforms and community forums
02Filtering prohibited content in real-time chat and collaboration tools
03Enforcing corporate compliance by managing custom lists of restricted terminology
