Autonomy Docs

Environments

Where a project runs, with a target per platform and the values its test cases use.

An environment is one place a project runs: Production, Staging, QA or a preview deployment. Test cases don't change between environments. A run picks an environment, and the environment supplies the URL, the app build and the values.

Open a project and click Environments. Owners and admins add and edit environments; members can see them.

Platform targets

An environment has one target for each platform it can test. A project tests every platform its environments have targets for.

PlatformTarget
WebA URL, such as https://qa.example.com. The agent uses Chrome.
APIA base URL. The agent sends HTTP requests directly, with no browser or device.
AndroidAn .apk or .aab you upload, or a direct download link to one.
iOSComing soon. Until iOS testing is available, an iOS target can't be saved.

Autonomy reads the app's package name or bundle ID, and its version, from the build: there's nothing to type. Runs use the build saved on the environment. To test a new build, upload it, or save the download link again in the environment. Uploads are checked for safety before a run can use them. To send a build from CI, see Mobile builds.

Environment values

Environment values are named values your steps use as {{env.NAME}}, such as {{env.TEST_USER_EMAIL}} or {{env.COUPON_CODE}}. Each run fills them in from the environment it runs in, so the same test case works against every environment.

  • Names are upper case, with underscores: TEST_USER_EMAIL.
  • Tick Secret for passwords, tokens and keys. Secrets are write-only: nobody can view a saved secret again, and only runs receive it. The agent fills a secret by reference and never sees its value, and it never appears in evidence.
  • A value can be overridden for one platform: Override a value for Android replaces the value with the same name on Android only.
  • If a step uses a name the environment doesn't define, the test case shows it as missing before you run it.

Values are read when a runner picks up a run. If you edit a value, queued runs use the new one. If you remove a value a queued run needs, that run ends as Couldn't start.

For accounts, use sign-ins instead of values: the agent picks the right account for a step and fills its password and authenticator code by reference.

Maintainers

Add maintainers to an environment to send them its failures, sign-in problems and weekly digest.

Archive an environment

Archiving an environment keeps its runs and issues. New runs can't pick it. Admins archive environments from the environment's page.

Production

When you run against an environment named Production, the agent behaves like a careful user: it won't spend real money or delete existing data unless a step asks for it. Use a test environment for anything destructive.

Didn’t find it?

On this page