Sweat Less, Test Better: Step MCP Is Here

What if your testers could leverage AI to more efficiently create, execute, analyze, and improve Step test suites?

With the release of Step 30 “Mont Blanc de Cheilon”, we’re shipping the first version of the Step MCP Server, bringing Step capabilities directly into AI-assisted development workflows, alongside broader improvements to analytics, execution reporting, integrations, and core platform capabilities.

Below you’ll find the highlights of the release. For the complete list of bug fixes and enhancements, visit the release notes.

Step MCP: AI-Assisted Testing, Natively

Whether you’re building with Microsoft Copilot, Claude Code, custom AI agents, or agent frameworks such as LangGraph, Step MCP enables your AI workflows to:

  • Create and validate Step-compliant automation packages
  • Execute tests locally or remotely at scale
  • Analyze results from any type of test, at any scale
  • Feed findings back into your DevOps or QA workflows
  • Prioritize next actions: fix the application, update tests, or refine specifications

Step MCP acts as a native bridge between AI agents and the Step platform, exposing years of testing and automation capabilities through a simple, AI-friendly interface.

Typical interactions

  • Add a login test for example.com to the test suite
  • Run the test suite in parallel on Step
  • Investigate potential failures and update the tests
  • Show me the final test code
  • Explain what happened during my last failed execution

Get started today

The Step MCP Server is available now as part of the Step 30 release.

Beyond AI

Step MCP is one of the most visible innovations in Step 30, but it’s only part of the story. This release also delivers major enhancements across analytics, reporting, integrations, and platform operations, including a new metrics framework, Prometheus improvements, customizable execution reports, and expanded Jira/Xray capabilities.

Key highlights

  • New metrics framework: keywords can now emit counters, gauges, and histograms alongside measurements, with execution- and controller-level metrics standardized across integrations
  • Prometheus improvements: support for native histograms and per-execution series cardinality
  • Customizable execution reports: fully configurable report layouts and a new contextual drill-down experience
  • Expanded Jira/Xray capabilities: reference existing Step plans from a Jira issue field, post-creation transitions, and variabilized issue-creation fields
  • Core platform: compatibility with Java 25 and granular access control for resources

📖 Explore everything that’s new in Step 30.

Read the full release notes
Categories: NEWS