The Agent Skills Directory

Official SDK Integration: The skill facilitates the use of the azure-ai-contentsafety package, which is an official library from a well-known service provider. It allows for the detection of various harm categories such as hate, sexual content, violence, and self-harm.\n- Secure Credential Handling: The documentation and code examples prioritize security by using environment variables for authentication. The inclusion of acceptance criteria that explicitly flag hardcoded secrets as an error reinforces secure development practices.\n- Resource Targeting: All network operations are directed towards official cloud service endpoints. The functionality for analyzing images via URLs is a standard feature of the integrated service, used for content moderation purposes.\n- Data Processing Context: While the skill processes user-generated text and images, it does so through a specialized moderation API designed to categorize and filter content, rather than executing it as code or passing it to unsafe internal functions.

azure-ai-contentsafety-py