Deploy Your App deploy-your-app

IMPORTANT
Adobe LLM Apps is currently in Beta.
Features, workflows, and UI shown here do not necessarily represent the final state of the product. To join the Beta, send an email to llm-apps-beta@adobe.com.

Once you have written your handler code and pushed it to your linked repository, you can deploy the app from the LLM Apps UI.

This is a shared step for every journey. After deployment, continue to test the ChatGPT plugin or test the Claude connector.

Start the deployment

Open the App Detail page and select Deploy.

Select the target environment, then select Deploy.

Deploy — select the target environment

The deployment runs four steps:

  1. Preparing — retrieves the configuration required to deploy the app.
  2. Start deployment — starts the background deployment process.
  3. Build app — installs dependencies and builds the latest repository code.
  4. Publish — publishes the app to Adobe I/O Runtime.

Deploy — deployment pipeline running

NOTE
If an action has metadata in the UI but no matching handler file in the repository, it is still registered. Invocations use a default stub handler until you add the real code.

After a successful deployment

When all steps complete, the dialog shows Deploy successful.

Deploy — successful deployment

Click Close to dismiss the dialog. Scroll down to the Test the app section on the App Detail page:

App Detail — copy the MCP server URL

Each deployed environment shows an MCP server URL. Select Copy URL and use it to create a plugin in the target LLM platform.

The Deployment history section shows the last 10 deployments:

Deployment history

Each row shows the target Environment (Stage or Production), Status (Success or Failed), and the Deployed at date. You can use this table to track when deployments happened and verify that the
latest deployment succeeded.

Next step

recommendation-more-help
llm-apps-help-main-toc