Claude Code

Configure Claude Code, the AI coding CLI by Anthropic, to route LLM requests through your agentgateway proxy running in Kubernetes.

Before you begin

1. Set up an agentgateway proxy.
2. Install the Claude Code CLI (npm install -g @anthropic-ai/claude-code).
3. Get an Anthropic API key from the Anthropic Console.

Get the gateway URL

export INGRESS_GW_ADDRESS=$(kubectl get svc -n agentgateway-system agentgateway-proxy \
  -o jsonpath='{.status.loadBalancer.ingress[0].ip}')

echo "Gateway address: $INGRESS_GW_ADDRESS"

export INGRESS_GW_ADDRESS=$(kubectl get svc -n agentgateway-system agentgateway-proxy \
  -o jsonpath='{.status.loadBalancer.ingress[0].hostname}')

echo "Gateway address: $INGRESS_GW_ADDRESS"

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

Set up the Anthropic backend

Create a secret, backend, and route to proxy Claude Code traffic through agentgateway.

1. Export your Anthropic API key.

   export ANTHROPIC_API_KEY="sk-ant-your-key-here"

2. Create a Kubernetes secret for your API key. For other authentication methods, see API keys.

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

3. Create an AgentgatewayBackend with the /v1/messages route and any other details such as models that you want to configure.

   Flexible model (recommended)Fixed model

   Allow Claude Code to use any model. The anthropic: {} syntax means no model is pinned.

   kubectl apply -f- <<EOF
   apiVersion: agentgateway.dev/v1alpha1
   kind: AgentgatewayBackend
   metadata:
     name: anthropic
     namespace: agentgateway-system
   spec:
     ai:
       provider:
         anthropic: {}
     policies:
       ai:
         routes:
           '/v1/messages': Messages
           '*': Passthrough
       auth:
         secretRef:
           name: anthropic-secret
   EOF

   Pin the backend to a specific model. The model must match what Claude Code is configured to use, because Claude Code sends the model name in API requests and agentgateway rejects mismatches.

   kubectl apply -f- <<EOF
   apiVersion: agentgateway.dev/v1alpha1
   kind: AgentgatewayBackend
   metadata:
     name: anthropic
     namespace: agentgateway-system
   spec:
     ai:
       provider:
         anthropic:
           model: claude-sonnet-4-5-20250929
     policies:
       ai:
         routes:
           '/v1/messages': Messages
           '*': Passthrough
       auth:
         secretRef:
           name: anthropic-secret
   EOF

4. Create an HTTPRoute to forward all traffic to the Anthropic backend. This route uses a / path prefix so that all requests, including /v1/messages and /v1/models, are forwarded to the backend.

   kubectl apply -f- <<EOF
   apiVersion: gateway.networking.k8s.io/v1
   kind: HTTPRoute
   metadata:
     name: claude
     namespace: agentgateway-system
   spec:
     parentRefs:
       - name: agentgateway-proxy
         namespace: agentgateway-system
     rules:
       - matches:
         - path:
             type: PathPrefix
             value: /
         backendRefs:
         - name: anthropic
           namespace: agentgateway-system
           group: agentgateway.dev
           kind: AgentgatewayBackend
   EOF

Configure Claude Code

Set the ANTHROPIC_BASE_URL environment variable to point Claude Code at your gateway address.

export ANTHROPIC_BASE_URL="http://$INGRESS_GW_ADDRESS"

kubectl port-forward -n agentgateway-system svc/agentgateway-proxy 8080:80 &
export ANTHROPIC_BASE_URL="http://localhost:8080"

Verify the connection

1. Send a single test prompt through agentgateway.

   claude -p "Hello"

2. Verify the request appears in the agentgateway proxy logs.

   kubectl logs deployment/agentgateway-proxy -n agentgateway-system --tail=5

   Example output:

   info  request gateway=agentgateway-system/agentgateway-proxy listener=http route=agentgateway-system/claude endpoint=api.anthropic.com:443 http.method=POST http.path=/v1/messages http.status=200 protocol=llm gen_ai.provider.name=anthropic gen_ai.request.model=claude-haiku-4-5-20251001 gen_ai.usage.input_tokens=14 gen_ai.usage.output_tokens=9 duration=706ms

3. Optionally, start Claude Code in interactive mode.

   claude

   All requests, including prompts, tool calls, and file reads, flow through agentgateway.

Next steps