Skip to main content
A Momentic test is a YAML file in your repository that mixes exact steps with goal-driven AI actions. npx momentic run executes a web test, and npx momentic-mobile run an iOS or Android test, on your machine and in CI. Mo’s Coder can write one from a Mo finding.
checkout.test.yaml
module runs a shared login sequence, the plain string is an AI action that plans its own clicks, and assert is a preset step that AI evaluates. For exploratory testing without a YAML file, use Mo.

Anatomy of a test

Step types

Use preset steps when the path needs exact, reviewable control, and AI actions when the goal matters more than the path. Preset steps still use AI to locate elements, evaluate assertions, or extract data, but the sequence stays explicit. See File format, Steps, Variables, and Modules for the authoring syntax. See AI action for goal-driven flows.

Code integration

Use preset steps for common interactions, then drop to code when a flow needs it:
  • JavaScript steps run custom code in a sandboxed Node environment or the current browser page.
  • appium steps execute an Appium script on the current device.
  • Modules and variables connect reusable setup, application state, and data across the flow.
See Test portability and migration for where your YAML lives and how to move a suite in or out.

Run locally and in CI

momentic run runs a file, a directory, or the whole suite, locally or in your CI system, with sharding for larger suites. AI test selection keeps pull-request feedback focused, while AI test maintenance turns eligible failures into recovery, diagnosis, and reviewable repairs.

CLI-based agents

The web CLI exposes these AI commands so they can be composed in CI, scripts, and custom agent loops: Each command reads repository files or saved results and can emit structured output, so your pipeline decides when it runs and what happens next. See the momentic ai reference and MCP server.

What carries across runs

  • The step cache reuses successful element resolutions for fast, deterministic replay.
  • Memory retrieves relevant decisions from earlier runs for locator and assertion agents.
  • The Knowledge base adds your terminology, rules, and known flows. Your team can customize it from the dashboard.