Analyzes text and images for harmful content using Azure AI Content Safety with custom blocklist support.
This skill integrates the Azure AI Content Safety REST SDK for TypeScript into Claude Code, enabling developers to automate content moderation for text and images. It provides built-in detection for categories like hate speech, sexual content, violence, and self-harm with granular severity scoring. Additionally, it supports advanced features like custom blocklist management and severity-based filtering, making it an essential tool for building safe, compliant, and user-centric applications within the Azure ecosystem.
Key Features
0131,722 GitHub stars
02Flexible image processing via Base64 content or Blob URLs
03Multi-category text and image analysis for hate, sexual, violence, and self-harm
04Granular severity scoring with 4-level and 8-level output types
05Type-safe moderation helpers with robust error handling via isUnexpected() guards
06Custom blocklist management for domain-specific prohibited terms
Use Cases
01Managing enterprise compliance through custom prohibited term blocklists
02Automating user-generated content moderation for community platforms
03Building safety filters for AI-generated text and visual outputs
