
When we first announced support for the Model Context Protocol (MCP) earlier this year, it was more than just another product update; it was a milestone for Enterprise Architecture.
For the first time, architects could securely connect conversational AI tools like Claude, ChatGPT, Gemini, and Microsoft Copilot to live architecture data, without the need for custom integrations or brittle connectors. The industry took notice. Analysts, customers, and peers alike recognized the significance: explainable, permission-aware AI, finally available to Enterprise Architecture.
Now, we’re moving from proving the concept to scaling it. Today, we’re proud to announce that the Ardoq AI Gateway (MCP Server) is moving from Beta to General Availability (GA).
Recap: What is AI Gateway (MCP Server)?

MCP (Model Context Protocol) is an open standard that acts as a structured bridge between AI and data platforms like Ardoq.
Instead of asking architects to manually export, clean, and share information, MCP lets AI tools query Ardoq directly and return contextual, permission-aware answers instantly and securely.
It’s the foundation of a future where EA isn’t about static documentation, but about real-time, AI-assisted decision-making.
What Our Technical Preview Delivered

The first iteration of AI Gateway (MCP Server) introduced a powerful but focused set of capabilities:
- Secure, Read-Only Access: Conversational AI could query Ardoq reports without compromising governance or data integrity.
- No Custom Integrations: AI assistants could connect directly through MCP—no brittle, one-off connectors required.
- Support for Leading AI Clients: It was tested and proven to work with Claude, Microsoft Copilot, ChatGPT, and other MCP-enabled tools.
- Foundation for Explainable AI in EA: Responses were based on live, governed data in Ardoq, not opaque black-box outputs.
This early preview phase proved the concept: AI could fetch and explain architecture data in plain English, safely and reliably.
What’s New in MCP GA

With GA, AI Gateway (MCP Server) takes a leap forward, moving from data retrieval to discovery.
- Full Viewpoint & Dashboard Access: Query not just reports but also dashboards and viewpoints across Ardoq.
- Metamodel Awareness: AI can now reason over your repository’s structure, enabling richer, context-aware questions.
- Exploratory Prompts: Move beyond “find me this data” to “what initiatives are tied to our finance objectives?” or “where do we have overlapping applications?”
- Ad-hoc Visualization (Client-Dependent): With supported AI clients, users can generate on-demand diagrams or visualizations from queried components via MCP.
- Enterprise-Ready Infrastructure: Logging, analytics, and accuracy evaluations provide production-scale reliability.
- Expanded AI Coverage: Tested and ready for Claude (Sonnet 4), ChatGPT (gpt-4o-mini and GPT-5 when available), Gemini (2.5 Pro), and Microsoft Copilot.
Together, these upgrades mean MCP doesn’t just fetch information. It helps you uncover inefficiencies, identify risks, and discover optimization opportunities, all through the AI tools you already use.
Why This Matters

The enterprise AI landscape is evolving rapidly. But without access to governed, structured data, even the most advanced assistants struggle to deliver meaningful business outcomes.
AI Gateway (MCP Server) fixes that. It ensures that conversational AI is not only powerful, but also trustworthy, explainable, and grounded in your enterprise architecture context.
- For architects, that means fewer hours spent retrieving, formatting, and sharing data.
- For business stakeholders, it means quick, accurate answers in familiar tools.
- For leadership, it means faster, clearer insights into inefficiencies and opportunities.
In short, AI Gateway (MCP Server) transforms Enterprise Architecture from a documentation-focused discipline into a discovery engine.
What Customers Are Saying

A Category-Defining First

No other Enterprise Architecture platform has brought MCP to GA. Competitors may have it on their roadmap, but Ardoq is first to market with secure, production-ready support.
And GA is not the finish line. It’s the launchpad.
AI Gateway (MCP Server) sets the stage for the next wave of explainable and easy-to-understand EA: AI that doesn’t just surface insights but reasons over options and recommends the next best action.
Learn More and See MCP in Action
With AI Gateway (MCP Server) GA, the future of Enterprise Architecture isn’t just imagined. It’s already here.