Momentic launches Mo, an AI agent that tests apps without scripts

Momentic built its business on helping developers describe browser test workflows in ordinary language, which a large language model can reduce the labor for. But the company ran into a deeper version of the same problem: even an easier-to-maintain test artifact is still something somebody has to maintain. Mo is presented as the evolution that removes that intermediate layer, letting QA agents work from requirements and intent directly. As co-founder and CEO Wei-Wei Wu put it, "As long as you have scripts somewhere in your codebase, someone has to maintain it."

Mo takes the shape of the AI-coding tools developers already use day to day, with a web interface and chat box plus a command-line interface, and it can pull context from a product requirements document, Jira issue, Linear ticket or Confluence document. The wager is that agentic testing scales where scripted suites do not, since the more complex an application becomes, the more tests are needed and the heavier the maintenance burden grows.

The stakes hinge on whether that bet holds for complex apps and on whether teams trust agent-generated results enough to drop their own suites. Momentic said customers including Notion have been putting Mo through its paces over the past few weeks, so early evidence from those deployments is likely to shape how quickly the script-free approach spreads.