Skip to main content
Mo is in private beta and may change. Sign up to get onboarded.
Mo needs no test suite and no code in your repository. Give it a running target and a brief. You will need:
  • Mo access (it is in private beta).
  • A running target: a URL, a preview deployment, or an uploaded mobile build. Staging is fine.
  • A test account, if the area sits behind a sign-in.
1

Open the composer

Open Mo in the Momentic app and select the target. Select Web and paste the exact URL, including the path of the area you want checked, or select an uploaded mobile build. A staging or preview URL is fine.
2

Write a narrow brief

Name one area, the account to sign in with, and the actions Mo must not take:
Mo asks you for anything it still needs before it starts. See write a good brief for the full field list.
3

Watch the run

Test cases and bugs appear while the session continues, and the first bugs can appear within minutes. Open the transcript of any sub-agent to see what it did, or message Mo to change the scope mid-run.
4

Read the report

The report holds a verdict per case and a bug entry for each failure a reproducer agent confirmed from a clean browser. See session reports.
5

Triage what it found

Accept a bug, or mark it as “works as intended” and give the reason. That reason goes to the knowledge base, so a later session does not report the same intended behavior. See triage.

​
Start a session from your coding agent

This is the durable path for most work: install the Mo skill once and ask the agent for the pass in plain language. Install the CLI first if you have not:
mo skills installs mo-qa into each detected agent. Without the mo CLI, use npx skills add https://github.com/momentic-ai/skills --skill mo-qa instead. Then invoke:
The skill installs and authenticates the mo CLI, writes the brief, starts the session, follows it, exports the report, and can fix and recheck the bugs it finds. See install the Mo skill, use the Mo CLI, and the mo CLI reference.

​
What a pass looks like

A file storage and sharing product ran Mo against staging before a release. The team wrote no tests for the pass.
  • The brief was the release scope and one staging URL.
  • The session agent planned the coverage and dispatched explore agents over the product areas in the scope.
  • At its widest the session held 116 agents at once, each with its own hosted browser.
  • 27 bugs reached the report, each one reproduced by a bug reproducer agent from a clean browser, with expected behavior, actual behavior, reproduction steps, and a recording.
  • The pass ran in an afternoon, and it covered work the team estimated at two days of manual QA.

​
Next

Run a session

Sign-in, test data endpoints, schedules, and what a pass costs.

Use cases

Pull requests, Vercel previews, local branches, and mobile builds.

How Mo works

The agents in a session, the targets, and the limits.