Setup & Installation
npx skills add https://github.com/microsoft/skills --skill azure-ai-contentsafety-ts
https://github.com/microsoft/skills/tree/main/.github/plugins/azure-sdk-typescript/skills/azure-ai-contentsafety-ts
View on GitHubWhat This Skill Does
Analyzes text and images for harmful content using Azure AI Content Safety. Detects hate speech, sexual content, violence, and self-harm across four severity levels. Supports custom blocklists for domain-specific term filtering.
It returns structured severity scores per category rather than a binary pass/fail, so you can apply different thresholds per context without building your own classifier.
When to use it
- Screening user-submitted forum posts for hate speech before publishing
- Flagging uploaded profile images that contain violent or sexual content
- Running custom blocklists against chat messages in a gaming platform
- Auditing support ticket text for self-harm language before routing to agents
- Applying severity thresholds to moderate product reviews at different risk levels