red teaming cloud update

Author: lgayhardt

articles/ai-foundry/how-to/develop/run-ai-red-teaming-cloud.md

@@ -7,7 +7,7 @@ ms.service: azure-ai-foundry
 ms.custom:
   - references_regions
 ms.topic: how-to
-ms.date: 06/03/2025
+ms.date: 07/23/2025
 ms.reviewer: minthigpen
 ms.author: lagayhar
 author: lgayhardt
@@ -31,7 +31,7 @@ If this is your first time running evaluations or AI red teaming runs on your Az
 
 ## Getting started
 
-First, install Azure AI Foundry SDK's project client which runs the AI Red Teaming Agent in the cloud
+First, install Azure AI Foundry SDK's project client, which runs the AI Red Teaming Agent in the cloud.
 
 ```python
 uv install azure-ai-projects azure-identity
@@ -46,15 +46,44 @@ Then, set your environment variables for your Azure AI Foundry resources
 import os
 
 endpoint = os.environ["PROJECT_ENDPOINT"] # Sample : https://<account_name>.services.ai.azure.com/api/projects/<project_name>
-model_endpoint = os.environ["MODEL_ENDPOINT"] # Sample : https://<account_name>.services.ai.azure.com
-model_api_key= os.environ["MODEL_API_KEY"]
-model_deployment_name = os.environ["MODEL_DEPLOYMENT_NAME"] # Sample : gpt-4o-mini
+
 ```
 
 ## Supported targets
 
 Running the AI Red Teaming Agent in the cloud currently only supports Azure OpenAI model deployments in your Azure AI Foundry project as a target.
 
+## Configure your target
+
+You can configure your target model deployment in two ways:
+
+### Option 1: Using Foundry project deployments
+
+If you're using model deployments that are part of your Azure AI Foundry project, set up the following environment variables:
+
+```python
+import os
+
+model_endpoint = os.environ["MODEL_ENDPOINT"] # Sample : https://<account_name>.openai.azure.com
+model_api_key = os.environ["MODEL_API_KEY"]
+model_deployment_name = os.environ["MODEL_DEPLOYMENT_NAME"] # Sample : gpt-4o-mini
+```
+
+### Option 2: Using Azure OpenAI/AI Services deployments
+
+If you want to use deployments from your Azure OpenAI or AI Services accounts, you first need to connect these resources to your Foundry project through connections.
+
+1. **Create a connection**: Follow the instructions in [Configure project connections](../configure-project-connection.md?pivots=ai-foundry-portal#add-a-connection) to connect your Azure OpenAI or AI Services resource to your Foundry project.
+
+2. **Get the connection name**: After connecting the account, you'll see the connection created with a generated name in your Foundry project.
+
+3. **Configure the target**: Use the format `"connectionName/deploymentName"` for your model deployment configuration:
+
+```python
+# Format: "connectionName/deploymentName"
+model_deployment_name = "my-openai-connection/gpt-4o-mini"
+```
+
 ## Create an AI red teaming run
 
 # [Python](#tab/python)
@@ -86,19 +115,28 @@ red_team_agent = RedTeam(
 )
 
 # Create and run the red teaming scan
-red_team_response = project_client.red_teams.create(red_team=red_team_agent, headers={"model-endpoint": model_endpoint, "api-key": model_api_key,})
+# If you configured target using Option 1, use:
+# headers = {"model-endpoint": model_endpoint, "api-key": model_api_key}
+# If you configured target using Option 2, use:
+# headers = {}
+
+# Choose one of the following based on your configuration option:
+headers = {"model-endpoint": model_endpoint, "api-key": model_api_key}  # For Option 1
+# headers = {}  # For Option 2
+
+red_team_response = project_client.red_teams.create(red_team=red_team_agent, headers=headers)
 ```
 
 # [cURL](#tab/curl)
 
 ```bash
-curl --request POST \  --url https://{{account}}.services.ai.azure.com/api/projects/{{project}}/redteams/runs:run \  --header 'content-type: application/json' \  --header 'authorization: Bearer {{ai_token}}'  --data '{  "scanName": "sample_scan_magic_1",  "riskCategories": [    "Violence"  ],  "attackStrategy": [    "Flip"  ],  "numTurns": 1,  "target": {    "type": "AzureOpenAIModel",    "modelDeploymentName": "{{connectionName}}/{{deploymentName}}"  }}'
+curl --request POST \  --url https://{{account}}.services.ai.azure.com/api/projects/{{project}}/redteams/runs:run \  --header 'content-type: application/json' \  --header 'authorization: Bearer {{ai_token}}'  --data '{  "displayName": "Red Team Scan #1",  "riskCategories": [ "Violence" ],  "attackStrategy": [ "Flip" ],  "numTurns": 1,  "target": {    "type": "AzureOpenAIModel",    "modelDeploymentName": "{{connectionName}}/{{deploymentName}}"  }}'
 ```
 
 - Replace `{{account}}`, `{{project}}` with Foundry Project account name and project name.
 - Replace `{{ai_token}}` with Bearer token with audience "<https://ai.azure.com>"
-- Replace `"{{connectionName}}"` with the Azure OpenAI model connection name connected to the Foundry project account.
-- Replace `"{{deploymentName}}"` with the Azure OpenAI deployment name of the AOAI connection account.
+- For Option 1 (Foundry project deployments): Replace `"{{connectionName}}/{{deploymentName}}"` with just `"{{deploymentName}}"` (your model deployment name).
+- For Option 2 (Azure OpenAI/AI Services deployments): Replace `"{{connectionName}}"` with the Azure OpenAI model connection name connected to the Foundry project account, and replace `"{{deploymentName}}"` with the Azure OpenAI deployment name of the Azure OpenAI connection account.
 
 ---