Sydney, Australia - 4 June 2026. Buzzy, the semantic application platform for governed AI app delivery, today announced the general availability of Buzzy Builder MCP, bringing governed enterprise app creation to MCP-enabled development environments including Codex, Claude Code, Cursor, and AI agents.
Buzzy has two MCP capabilities. Buzzy Custom MCP, already in market, lets applications built with Buzzy expose their data, functions, and workflows to MCP-enabled AI assistants through governed interfaces. Buzzy Builder MCP, the new capability being released, brings Model Context Protocol into the app creation workflow itself, allowing AI tools and agents to help generate and refine the semantic app definitions that power Buzzy applications.
Each Buzzy application has its own semantic app definition, capturing the application's intent, brief, flows, blueprint, data model, privacy settings, UI, logic, security requirements, and deployment behaviour. These many app definitions run on one maintained Buzzy core engine, which generates production-ready web and native mobile applications.
AI app development is moving beyond prompt-to-code. The next wave is prompt-to-structure: AI helping define the application clearly enough that a trusted platform can build it, run it, test it, secure it, and govern it. Buzzy's semantic app definition gives enterprises that control point. With Buzzy Builder MCP, teams can use the AI tools they already love while still producing applications that are structured, governable, and maintainable.
Adam Ginsburg, Founder and CEO of Buzzy
The launch comes as AI-assisted software development moves rapidly into the enterprise. Gartner has predicted that by 2028, 75% of enterprise software engineers will use AI code assistants, up from less than 10% in early 2023. But faster code creation is also creating new risks. Veracode's 2025 GenAI Code Security Report found that AI-generated code introduced risky security flaws in 45% of tests, while Harness reported that heavy AI coding tool users deploy faster but also experience more frequent deployment problems when AI-generated code is involved.
This is the problem Buzzy is built to address. AI code generation can create speed, but it can also create code sprawl, duplicated logic, inconsistent security models, fragile integrations, and maintenance debt. Each generated application can become another codebase to audit, patch, refactor, test, deploy, and maintain.
Buzzy takes a different path. Instead of asking AI to generate large, arbitrary codebases, Buzzy uses AI to help create structured semantic app definitions. Because the application is defined in a semantic layer, Buzzy can use that same definition to help drive testing, privacy controls, security review, compliance checks, and promotion across development, staging, and production environments.
Field-level privacy, testing, security review, and agents
The release also includes field-level privacy controls, now generally available, giving teams explicit control over sensitive data across runtime, publications, editor views, and REST/MCP paths. Automated testing and the App Security Review Scanner are now in beta, helping teams record repeatable tests, verify data model access checks, identify application security requirements, and apply fixes inside the editor.
Buzzy also supports enterprise UI toolkits and reusable business cookbooks, allowing teams to align AI-generated applications with approved design systems, internal standards, and business patterns. Buzzy Agents, coming soon, will extend this governed model further by aligning agents with the Buzzy app definition, security model, and compliance requirements.
Enterprises do not just need AI that can produce a demo. They need AI that can help produce a governed application definition - one that can be inspected, secured, tested, promoted through environments, and maintained over time. That is the difference between creating another codebase and creating an application the business can trust.
Adam Ginsburg, Founder and CEO of Buzzy
Buzzy Builder MCP is now generally available. Field-level privacy controls are also generally available. Automated testing and security review are in beta, with Buzzy Agents to follow soon.
Teams interested in access can visit buzzy.buzz and join the waitlist.
About Buzzy
Buzzy is a semantic application platform for governed AI app delivery. Buzzy uses AI to help create semantic app definitions that run on a maintained core engine, delivering production-ready web and native mobile applications with less technical debt, stronger governance, and a more manageable application lifecycle.