Azure AI Foundry

Route LLM requests to Azure AI Foundry (Azure OpenAI) through agentgateway on Kubernetes, using custom path routing and URL rewriting.

What you’ll build

In this tutorial, you will:

Set up a local Kubernetes cluster with agentgateway
Configure an Azure OpenAI backend with Azure AI Foundry credentials
Set up path-based routing with URL rewriting
Send chat completion requests through agentgateway to Azure OpenAI

Before you begin

Make sure you have the following tools installed:

You also need:

An Azure account with access to Azure AI Foundry
An Azure OpenAI deployment with an API key
Your Azure OpenAI endpoint URL (e.g., your-resource.services.ai.azure.com)
Your deployment name (e.g., gpt-4o-mini)

For detailed tool installation instructions, see the LLM Gateway tutorial.

Step 1: Create a kind cluster

kind create cluster --name agentgateway

Step 2: Install agentgateway

# Gateway API CRDs
kubectl apply -f https://github.com/kubernetes-sigs/gateway-api/releases/download/v1.4.0/standard-install.yaml

# agentgateway CRDs
helm upgrade -i --create-namespace \
  --namespace agentgateway-system \
  --version v2.3.0-main agentgateway-crds oci://ghcr.io/kgateway-dev/charts/agentgateway-crds

# Control plane
helm upgrade -i -n agentgateway-system agentgateway oci://ghcr.io/kgateway-dev/charts/agentgateway \
  --version v2.3.0-main

Step 3: Create a Gateway

kubectl apply -f- <<EOF
apiVersion: gateway.networking.k8s.io/v1
kind: Gateway
metadata:
  name: agentgateway-proxy
  namespace: agentgateway-system
spec:
  gatewayClassName: agentgateway
  listeners:
  - protocol: HTTP
    port: 80
    name: http
    allowedRoutes:
      namespaces:
        from: All
EOF

Wait for the proxy:

kubectl get deployment agentgateway-proxy -n agentgateway-system

Step 4: Set your Azure credentials

Export your Azure AI Foundry API key:

export AZURE_FOUNDRY_API_KEY=<insert your Azure API key>

Step 5: Create the Kubernetes secret

Store your Azure credentials in a Kubernetes secret:

kubectl apply -f- <<EOF
apiVersion: v1
kind: Secret
metadata:
  name: azureopenai-secret
  namespace: agentgateway-system
type: Opaque
stringData:
  Authorization: $AZURE_FOUNDRY_API_KEY
EOF

Step 6: Create the Azure OpenAI backend

Create an AgentgatewayBackend resource configured for Azure OpenAI. Replace your-resource.services.ai.azure.com with your actual Azure AI Foundry endpoint, and gpt-4o-mini with your deployment name.

kubectl apply -f- <<EOF
apiVersion: agentgateway.dev/v1alpha1
kind: AgentgatewayBackend
metadata:
  name: azureopenai
  namespace: agentgateway-system
spec:
  ai:
    provider:
      azureopenai:
        endpoint: your-resource.services.ai.azure.com
        deploymentName: gpt-4o-mini
        apiVersion: 2025-01-01-preview
  policies:
    auth:
      secretRef:
        name: azureopenai-secret
EOF

Setting	Description
`azureopenai.endpoint`	Your Azure AI Foundry resource endpoint
`azureopenai.deploymentName`	The name of your Azure OpenAI model deployment
`azureopenai.apiVersion`	The Azure OpenAI API version to use
`policies.auth.secretRef`	Reference to the secret containing your API key

Verify the backend was created:

kubectl get agentgatewaybackend -n agentgateway-system

Step 7: Create the HTTPRoute with URL rewriting

Create an HTTPRoute that routes requests on the /azureopenai path and rewrites the URL to the chat completions endpoint.

kubectl apply -f- <<EOF
apiVersion: gateway.networking.k8s.io/v1
kind: HTTPRoute
metadata:
  name: azureopenai
  namespace: agentgateway-system
spec:
  parentRefs:
    - name: agentgateway-proxy
      namespace: agentgateway-system
  rules:
  - matches:
    - path:
        type: PathPrefix
        value: /azureopenai
    filters:
    - type: URLRewrite
      urlRewrite:
        path:
          type: ReplaceFullPath
          replaceFullPath: /v1/chat/completions
    backendRefs:
    - name: azureopenai
      namespace: agentgateway-system
      group: agentgateway.dev
      kind: AgentgatewayBackend
EOF

This route:

Matches requests to /azureopenai
Rewrites the path to /v1/chat/completions before forwarding to Azure
Routes to the Azure OpenAI backend

Step 8: Test the API

Set up port-forwarding:

kubectl port-forward deployment/agentgateway-proxy -n agentgateway-system 8080:80 &

Send a request to Azure OpenAI through agentgateway:

curl "localhost:8080/azureopenai" \
  -H "Content-Type: application/json" \
  -d '{
    "messages": [
      {
        "role": "system",
        "content": "You are a helpful assistant."
      },
      {
        "role": "user",
        "content": "What is Azure AI Foundry in one sentence?"
      }
    ]
  }' | jq

Example output:

{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "choices": [{
    "message": {
      "role": "assistant",
      "content": "Azure AI Foundry is Microsoft's unified platform for building, deploying, and managing AI applications and models at scale."
    },
    "index": 0,
    "finish_reason": "stop"
  }]
}

Multiple Azure deployments

You can route to different Azure OpenAI deployments based on request paths. Create additional backends and routes for each deployment:

# Backend for a different model
apiVersion: agentgateway.dev/v1alpha1
kind: AgentgatewayBackend
metadata:
  name: azureopenai-gpt4
  namespace: agentgateway-system
spec:
  ai:
    provider:
      azureopenai:
        endpoint: your-resource.services.ai.azure.com
        deploymentName: gpt-4
        apiVersion: 2025-01-01-preview
  policies:
    auth:
      secretRef:
        name: azureopenai-secret

Then add an HTTPRoute matching /azureopenai-gpt4 to route to the new backend.

Cleanup

kill %1 2>/dev/null
kind delete cluster --name agentgateway

Next steps

Azure OpenAI Reference

Complete Azure OpenAI configuration

LLM Gateway

Route to multiple LLM providers

AI Prompt Guard

Protect your LLM requests