Chat assistant
In this introductory tutorial, you’ll use the Writer Framework to build an AI chat assistant.
Setting up your project
Creating a Writer app and getting your API key
From the Home screen, click on Build an app.
Select Framework as the app type you’d like to create, enabling you to generate keys and build your app with the Writer Framework.
On the next screen, you can edit your Writer application name in the upper left. Underneath “Authenticate with an API key,” click on Reveal to see and copy your API key.
Creating the application
Next, open your terminal and navigate to the directory where you want to create your application directory.
Set the API key environment variable
To pass your API key to the Writer Framework, you need to set an environment variable called WRITER_API_KEY. One simple way to do this is by exporting the variable for your terminal session.
Create the application
Run the following command to create your application, replacing chat-assistant with your desired project name and ai-starter with the template you want to use:
writer create chat-assistant --template=ai-starter
This command sets up a new project called chat-assistant in the specified directory.
Edit your project
To edit your project, run the below commands. This will bring up the console, where Framework-wide messages and errors will appear, including logs from the API. By default, the Writer Framework Builder is accessible at localhost:3006. If that port is in use, you can specify a different port. Open this address in your browser to view your default application setup.
Creating the UI
The Writer Framework lets you set up any layout according to your preferences with a fast drag-and-drop UI.
To rename this application and update the Header component, open the code editor and update my_app.title in wf.init_state:
wf.init_state({
"my_app": {
"title": "CHAT ASSISTANT"
}
})
Click the provided Section component to open its Component settings and clear out the default title. If you’d like to provide any other instructions or context to the user, you can also drag a Text component into the section.
Finally, drag a Chatbot component into the Section beneath the Text box.
Updating the code
With the UI built, you can now update your code to add chat functionality.
Initialize your application state
First, clear any default-generated state and add a conversation property set to writer.ai.Conversation(). Update your initial_state as follows:
wf.init_state({
// Other state elements above
"conversation": writer.ai.Conversation()
})
The Conversation method can optionally accept a dictionary or a content prompt (e.g., “You are a social media expert in the financial services industry”), but it can also be left empty to use the defaults.
Create a handler for the Chatbot component
Next, create a handler for incoming messages by adding the handle_simple_message handler. This method will manage the chat interactions:
def handle_simple_message(state, payload):
state["conversation"] += payload
for chunk in state["conversation"].stream_complete():
state["conversation"] += chunk
This code uses the streaming function of the Conversation method, which is a wrapper for the chat API endpoint. Each chunk returned from the stream is added to the conversation variable in the application state.
Binding to the UI
Click on the chatbot component to open up the Component settings panel. Bind this chatbot to a conversation variable by adding @{conversation} in the Conversation Object property under General. This variable will reference the Writer AI SDK. You can also update properties such as the assistant’s initials, the user’s initials, and whether the chat uses markdown.
Finally, attach the handler to the chatbot. In the User Interface, click on the chatbot component to bring up the Component settings panel. Scroll to the Events section towards the bottom of the pane and choose the handle_simple_message handler for the wf-chatbot-message event.
After saving and running your code, click the preview button and type something into your chat assistant. You should see the response appear on the screen as it comes back from the assistant. Congratulations!
Deploying the application
To deploy the application to the Writer cloud, either terminate your current Writer Framework process or open a new terminal session and run the following command:
writer deploy chat-assistant
Once the application is deployed, the CLI will return with the URL of your live application.
Conclusion
That’s all it takes to set up a basic application with the Writer Framework. This setup not only demonstrates the platform’s capabilities, but also provides a foundation on which you can build more complex applications. To learn more, explore the rest of the Writer Framework documentation and the API documentation.