Govern Your AI Agents on Azure: AKS, AI Foundry, Workload Identity #1012
Closed
Knapp-Kevin
started this conversation in
Show and tell
Replies: 1 comment
-
|
Great writeup @Knapp-Kevin — thank you for putting this together! The Container Apps sidecar pattern and Workload Identity bridge are exactly the deployment model we had in mind. We recently shipped the OpenClaw sidecar Helm chart and Dockerfile which aligns well with this guide. Would love your feedback on whether the sidecar image works smoothly in your AKS setup. This is a valuable community resource — we will link to it from our deployment docs. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Govern Your AI Agents on Azure: Container Apps, AI Foundry, Workload Identity
The Agent Governance Toolkit was built on Azure. Here's what's available today.
🏗️ Deploy on Container Apps in 4 Steps
The governance toolkit deploys as a sidecar container alongside your agent in a Container Apps Environment:
Full environment setup, Log Analytics workspace wiring, and the sidecar YAML manifest in the Azure Container Apps deployment guide.
🤖 Govern Azure AI Foundry Agents
AGT works with Foundry Agent Service. Wrap tool calls with
PolicyEvaluator.evaluate()and every action goes through your YAML policy engine before it runs:🔑 Workload Identity ↔ DID Bridge
EntraManagedIdentitymaps AGT's Ed25519 agent DIDs to user-assigned managed identities via federated credentials. Your governed agents acquire Azure tokens using their mesh identity. No client secrets, no key files:📊 Enterprise Azure Stack
📚 Full Docs
pip install agentmesh-platformAGT is pure Python, MIT licensed, and vendor-independent. Azure's where it was born, but it runs identically on AWS and GCP. 🚀
Beta Was this translation helpful? Give feedback.
All reactions