Azure AI Content Safety
Run Azure AI Content Safety as a safety provider inside DeepintShield policies.
Use Azure AI Content Safety as an external safety provider, linked to DeepintShield traffic through a guardrail policy.
What works
- Native adapter, linked through a policy bound to Virtual Keys
- Parallel execution with other safety providers for one policy
- Live MCP tool-call evaluation
- Outcomes of allow, allow with redaction, deny, sandbox or human approval
- A Test button that sends a live probe
Limits to know: Safety providers need the Team plan or above. A provider policy has a 10-second timeout.
How to connect
Add Azure AI Content Safety under safety providers in the console with your Azure details, test it, then link it to a guardrail policy.
Read the safety providers docs. Safety providers sit alongside the built-in checks in Runtime Guardrails.