MCP Server v0.12.0 is live — what’s new and why it matters

Hello MCP community,

We just pushed the v0.12.0 release of SmartBear MCP Server — a solid step forward in adding new capabilities across our ecosystem of products. 

What’s new in v0.12.0

• Zephyr support improvements — we added two new tools: one to retrieve Environments, and another to fetch a list of Test Executions. 
• Swagger enhancements — two powerful tools added:
  • create_api_from_prompt: generate API definitions directly from natural-language descriptions, powered by SmartBear AI, with governance and standardization applied. 
  • standardize_api: scan an existing API definition and automatically fix it to comply with standardization rules — helpful for teams managing many APIs or enforcing policies. 
• BugSnag observability extension — added tools for querying performance data and managing network-grouping rules: list/get span groups, list spans, get traces, list trace fields, and get/set network endpoint groupings. 
• QMetry automation & context metadata improvements — new tools for importing automation results, creating releases/cycles, and improved contextual metadata to help with LLM-based prompt handling. 
• Misc tools & UX improvements — support for a --version command line flag to easily check which MCP version you’re running. 
• Breaking/Changed Behavior in BugSnag integration — event threads have been removed from the “Get Error” response; instead, there is now “Get Event (by ID)”. If your automation or scripts rely on the old shape, you’ll need to update. 

Why this matters

• Teams using Zephyr now have deeper visibility into environments and test executions directly from MCP clients. That’s a big win for QA workflows and auditability.
• For API-first teams, the new Swagger AI-powered API creation and standardization tools could drastically speed up API lifecycle: from “idea” to “compliant spec” in one step — helping enforce consistency and governance at scale.
• The expanded BugSnag performance/tracing support bridges the gap between observability and AI-driven diagnostics. You can now query traces, spans, and network groupings — which makes MCP a more compelling tool for real-time debugging and error triage.
• Improved QMetry automation import + LLM context metadata makes it more practical to embed test results and context into AI-powered workflows — good for release automation, reporting, or even automated triage.
• The release overall tightens integration across the SmartBear ecosystem — pushing farther toward a unified, AI-ready platform.

What to do next

• Upgrade to v0.12.0. If you use BugSnag integrations — especially error queries — verify your scripts against the changed “Get Error / Get Event” shape.
• Try out the new Swagger and Zephyr tools. Think about how these could reshape your API and QA workflows.
• If you’re experimenting with AI-driven workflows (e.g. auto-generating APIs or linking runtime errors to API docs) now is a good time to test and give feedback.
• As always: share your experiences. What works? What needs polish? Your feedback will guide what’s next.

This is a big release with real substance behind it. For teams already invested in BugSnag, Swagger, Zephyr or QMetry, adopting v0.12.0 could unlock real productivity and automation opportunities.