Develop a C3 AI Application
You can develop applications on the C3 Agentic AI Platform using the C3 AI VS Code extension and C3 AI Studio.
Before creating a basic application, complete the requirements and prerequisites. See Prerequisites for C3 AI Application development.
Here's how to create a basic application:
- Connect to a single-node environment (SNE)
- Start your application
- Develop your application by creating a custom Type and methods
- Test your application with a preferred framework
Connect to a single node environment
Before connecting to a SNE in the C3 AI VS Code Extension, make sure there are no running synchronization processes. Do not remove or delete files during synchronization to maintain system stability and prevent potential issues.
Launch C3 AI VS Code Extension.
Select BROWSE TO ADD WORKSPACE.
Select the directory one level above your package. For example, if your workspace looks like the sample shown in Prerequisites for C3 AI Application development, select the
LocalWorkspacefolder.In the Environment URL field, enter your environment address. You can find this value in the Environments list of C3 AI Studio. Select the ellipsis next to your environment, then Open in VS Code.
- Alternatively, select Open VS Code in browser to open the C3 AI Extension in your browser.
Select CONNECT. You can view the OUTPUT tab to monitor the connection and synchronization process.
Authenticate your connection when prompted. Follow the prompts to open VS Code and complete authentication. This step is mandatory for first-time users, but may not apply to you if you have already successfully logged in.
The synchronization process begins and you can monitor the status appear the notification box of the C3 AI VS Code Extension. If you do not see the notification, select the bell icon at the bottom right corner.
A success notification indicates that the C3 AI VS Code Extension has connected to your environment.
Review validation
After the synchronization process, the remote environment validates all the new files that it received. During this process, the following status icons display next to the package files:
| Icon | Status | Description |
|---|---|---|
 | Unsynced | The local file has not been upserted to the server. This often indicates unsaved changes. |
 | Synced | The local file has been upserted to the server. |
 | Validated | The file has successfully synced to and validated by the remote environment. If the file is synced and validated but has errors, the file name turns red. |
Troubleshoot connecting to a single node environment
If you're unable to select a workspace, make sure the directory that contains your package does not use any special characters, spaces, or periods in the workspace path.
If you see the error message Unable to connect to C3 environment <environment link> is not a valid or active C3 Environment URL in the OUTPUT tab, verify if you are connecting to the correct environment.
When you connect to an environment for the first time, your package size and the strength of your wireless network determines the initial synchronization time. An initial synchronization with a large application package can take up to 10 minutes. If the connection takes more than 20 minutes, exit the VS Code application and re-connect. Subsequent connections can be faster.
If a file name turns red during the synchronization process, see the OUTPUT tab to view the errors and troubleshoot accordingly.
Start and view your application
Right-click the application package and select Start App. This initiates your application. A green circle icon next to the application name indicates that it is running.
After the application starts, you can view your application and other information like application errors in C3 AI Studio.
To learn more about using C3 AI Studio, see Introduction to C3 AI Studio
Develop your application
Now that you have a basic application, you can develop it further by creating custom methods and using more C3 AI VS Code Extension features. See Create Custom Types and Methods to learn more about further developing your application. For information on adding an application UI, see Create a Full-Stack Application Package.
Test your application
Create tests to confirm that your Types and application code perform as expected. C3 AI applications support popular unit and integration testing frameworks such as Jasmine and Python. For information on supported frameworks, see the C3 AI Testing Guide. For guidance on using Jasmine or Pytest, visit the following documentation: