mabl Unveils Next-generation Agentic Testing Platform for the AI Development Era

Release delivers what mabl calls “Active Coverage” — quality validation built to keep pace with AI coding agents — addressing a gap revealed in the company’s 2026 State of Quality Engineering Report.

mabl, the agentic testing platform built for enterprise teams, released a number of new capabilities designed to enable continuous quality at agentic development speed. As the speed and scale of AI-generated code continue to skyrocket within organizations in every industry, “Active Coverage” has become imperative, according to the company, which has been AI-native since 2017.

The launch coincides with findings from mabl’s 2026 State of Quality Engineering Report, a survey of nearly 1,000 software professionals that confirms the gap between code generation velocity and quality validation is widening. Among teams using AI coding agents today, the report found a near-even split: 41% say AI has improved code quality, while 37% say it has produced code faster but at lower quality, underscoring how much a team’s quality foundation determines which side of that gap they land on.

“Coding agents ship faster than any team in software history, but an agent grading its own work is biased toward shipping,” said Dan Belcher, Co-founder of mabl. Our enterprise customers told us the quality layer has to be independent, and it has to keep up,” We built mabl to be that layer, an agentic quality platform that tests both hand-crafted and AI-generated code with the rigor of an independent reviewer, and the observability, lineage, and retention the enterprise has always needed.”

Also Read: AiThority Interview with Glenn Jocher, Founder & CEO, Ultralytics

mabl’s new capabilities were designed to deliver Active Coverage in support of a new era of agentic software development:

• Agent Instructions makes team quality standards persistent and self-enforcing, encoding application-specific context directly into mabl so it’s applied automatically across every test it authors, every failure it analyzes, and every recovery it attempts.
• Cloud Test Generation allows tests to be authored entirely in the cloud with no local setup required, triggered from a browser, CLI, or IDE, with multiple sessions running in parallel so coverage keeps pace with development without creating a bottleneck.
• Runtime Recovery autonomously resolves unexpected obstacles during test execution, keeping tests running through environmental noise that would otherwise stop a pipeline cold — so when the pipeline stops, something actually broke.
• Conversational Results Analysis lets engineers interrogate test runs through natural language across individual tests, deployments, or the full workspace, turning hours of manual log investigation into minutes.
• Atlassian Rovo Integration brings mabl’s testing intelligence directly into Jira and Confluence, so teams can trigger runs, investigate failures, and assess release readiness without leaving the tools they already work in.