The correct option is Keep reusable non sensitive configuration outside the code.

Defining environment variables lets the team supply values like the API base URL and the logging level per environment without changing application code. This approach centralizes configuration for staging and production, avoids duplication, and keeps the codebase clean while still allowing different values in different Google Cloud projects.

Because these values are not secrets, storing them as environment variables in the workflow or on the Cloud Run service is appropriate. The workflow can set or pass variables for each environment and Cloud Run can inject them into the container at runtime, which results in predictable and repeatable deployments.

Control application logic and branching at runtime is not the best match for this scenario since the goal is to externalize environment specific configuration rather than to drive feature flags or conditional flow. GitHub Actions already provides dedicated conditional expressions for workflow branching and the application should remain focused on reading configuration rather than deciding control flow based on it.

Configure Cloud Run CPU and memory limits is incorrect because those limits are part of the Cloud Run service configuration and are set with deployment flags or service settings rather than with application environment variables.

Set absolute file paths for data input and output is incorrect because containers on Cloud Run should not rely on absolute host paths and the filesystem is ephemeral. If file locations must be configured, it is better to reference external services such as Cloud Storage or use relative paths within the container, and this is unrelated to the need to keep environment specific configuration out of the code.

Full AWS Practitioner Certification Question