Updated on 2026-10-23 GMT+08:00

Practice Process

Creating a Project

  1. Log in to CodeArts Agent IDE using your account and send a requirement to AI:

    Help me create a Scrum project named travel-assistant under my current account, to be used for developing a travel assistant.

  2. After the project is created, ask AI to create an epic:

    Create an epic named "Travel Assistant v1.0" under this project.

  3. After the epic is created, ask AI to split the epic into features:

    Split this epic into these features: Scenic Spot Recommendation, Itinerary Planning, Weather Query, Translation Services, and User Management.

    After AI successfully splits the epic into features, go to the CodeArts Req page to check the result. If the result is not as expected, ask the AI to fix it right away.

Creating a Repository/Code Check Rule

  1. Send a request to AI to create a repository for subsequent code development.

    Create a Python repository named "travel-assistant" in CodeArts Repo.

  2. Ask AI to create a code check task for the repository. The task will be automatically triggered when code is committed.

    Create a code check task for the code repository. Code check is automatically triggered when code is committed.

  3. Send a request to AI for developing the code for the preceding features.

    Develop the code for these features separately and complete self-verification.

  4. After the code development is complete, push the code.

    Push the developed code to the code repository "travel-assistant."

    When the code is pushed, the code check task is triggered.

Modifying and Merging Code

  1. Ask AI to check whether the code check task is successfully executed.

    Check whether the code check task is successful. If it fails, fix the issues and re-execute the check task until it is successful.

  2. Modify the code based on the code check task.

    View the details and fixing suggestions of major defects. Fix the code based on the suggestions and push the code again until it passes the check.

Building an Application and Pushing Its Image to SWR

  1. Ask AI to build a task.

    Create a build task, build a Docker image for the travel assistant, and push the image to SWR.

  2. After the build task is created, trigger the build.

    Trigger the build.

  3. After the build is complete, ask AI to check the build status.

    Check the build status. If the build fails, analyze the build logs until the build is successful.

Deploying the Assistant to CCE

  1. Ask AI to deploy the assistant to a CCE cluster.

    Deploy the travel assistant to a CCE cluster. If the deployment fails, analyze the deployment logs until the deployment is successful.

  2. After the deployment is successful, verify the services.

    Check whether the services are normal.

Managing Project Tests

  1. Ask AI to create a test version and plan.

    Create a test version and plan.

  2. Ask AI to write test cases.

    Create test cases for the travel assistant to cover all modules.

  3. Ask AI to create a test suite based on the test cases.

    Create a manual test suite, add the test cases created in the previous step to the suite, and execute the test.

  4. Check the test cases that failed.

    Check the failed test cases, locate the failure causes, and fix the issues.

Configuring a Pipeline

  1. Ask AI to configure a pipeline that is automatically triggered when code is pushed.

    Configure a pipeline that connects code check, build and push, and deployment, and is automatically triggered when code is pushed.

  2. Ask AI to run the pipeline and check whether it can run properly.

    Run the pipeline and return the pipeline running status.

  3. (Optional) If the pipeline runs successfully, skip this step and proceed to the next phase to release resources.

    If the pipeline fails to run, ask AI to check and fix issues.

    Analyze the pipeline run failure logs and fix the issues until the pipeline can run successfully.

Releasing Resources

This case involves pay-per-use CCE and ECS resources. If you no longer need these resources after completing this case, you are advised to release them promptly. Deleted resources cannot be restored.

Ask AI to clear all resources.
Clear all resources of the travel assistant, including the pipeline, build task, check task, deployed application, and test resources.

Clear all resources in reverse order based on the dependencies prompted by AI.