SAN FRANCISCO, CA – Momentic has launched a major platform update that reimagines software validation for the AI era. AI coding tools have made it possible for teams of all sizes to code more than ever before, but how that code is verified hasn’t kept up. The result is bugs arriving in production faster than teams can handle, and engineering teams holding fire-fighting hotfixes instead of shipping them.
Every Temporary Agent – whether it’s creating new tests, breaking down failures, or healing broken ones – works on the basis of shared knowledge of how your product really works. Groups can define terms, critical flows, and common failure patterns; Momentic maps the entire user journey and maintains that understanding as the product evolves. It imports Jira tickets, GitHub, support documents, Slack, and connects with other internal tools, so the system always knows what’s changing, what’s breaking, and whether the failure is a real bug or from a targeted change. The result is a system that gets smarter the more you use it.
Key capabilities in this release include:
Testing Installations That Grow With Your Product
Many teams share the same problem: coverage gaps that extend throughout the sprint because writing new tests doesn’t keep pace with the speed of the ship. Momentic’s Explore Agent closes that gap automatically. It monitors pull requests, reads diffs and codebase content, identifies which flows are missing, and suggests new tests or updates existing ones, without manual approval.
Overcoming Isolation and Healing: From Noise to Signal
Rigid testing and noisy CI pipelines are one of the biggest drains on engineering time, and one of the main reasons teams stop relying on their test rigs. The Momentian Failure Classification Agent automatically classifies all failures: a real fault, a target system change, a test setup problem, or a momentary fault. If the failure is not a real bug, Momentic suggests a fix and opens a PR. If it does, the team receives a high signal alert with full context of what is broken and why.
Legacy test frameworks produce artifacts that can only be understood by the developer who wrote them. Momentic’s new test format is purpose-based and readable by both humans and AI agents. Developers describe what they want to test in plain English; Momentic does it live against a browser or emulator. AI agents can analyze, build, and modify tests more effectively, making the entire development loop faster and more independent.
Momentic customers are already seeing the benefits of its measurable impact releases. In just a few weeks of beta access, Momentic agents analyzed over 70,000 test failures, opened 400 pull requests, and wrote 600 tests for a 73% integration rate. In total, customers have taken more than 2 billion steps on the platform and verified more than 80,000 PRs: proof that agent verification is not only compatible with AI coding tools, it actually gives developers the confidence to ship more with them, quickly.
For more information on Momentic availability and pricing, visit momic.ai/pricing.
