Hi Brandon, GM, I do not have Postman account. I just use the regular js to call the API. My code are attached. The workflow is: Project Setup (if not already done): my-chatbot-project/ ├── server.js <-- The Node.js backend code ├── package.json <-- Generated by `npm init` ├── node_modules/ <-- Dependencies installed here └── public/ ├── index.html <-- Your HTML file ├── style.css <-- Your CSS file (if any) └── script.js <-- The updated browser-side JavaScript Install Node.js Dependencies: Open your terminal, navigate to your my-chatbot-project directory, and run: Bash npm install express @google-cloud/dialogflow-cx Configure server.js: Open server.js and replace "your-gcp-project-id", "your-vertex-ai-agent-id" with your actual Google Cloud Project ID and the ID of your deployed Vertex AI Agent. Crucial for Authentication: Ensure your Node.js environment is authenticated to access Google Cloud. The easiest way for local development is to run gcloud auth application-default login in your terminal. For production, you'd typically use a service account key file via the GOOGLE_APPLICATION_CREDENTIALS environment variable. Place Frontend Files: Make sure your index.html, style.css (if any), and the updated script.js are placed inside the public/ directory. Start the Backend Server: In your terminal (still in your project directory), run: Bash node server.js You should see confirmation messages like: Backend server listening at http://localhost:3000 Frontend accessible via http://localhost:3000/index.html Access in Browser: Open your web browser and go to http://localhost:3000/index.html (or http://localhost:3000 if your server config serves index.html by default). Now, your chatbot should correctly send messages to your Node.js backend, which will then use the Dialogflow CX client to interact with your Vertex AI Agent, and stream the responses back. This setup provides the secure and recommended way to integrate with Vertex AI Agent Engine APIs from a browser-based frontend.

Hi @Tom Welsh Thanks a lot for your comment. Please see my above reply comment.

Thank you very much! Brandon. I will attend your class afternoon.