Introducing CData MCP Servers: Enterprise Data Access for Your AI, Finally

Generative AI has revolutionized our lives by applying its capabilities to public data. But it has been slow to bring that revolution to the workplace because of the difficulty of connecting LLMs to enterprise data.

Until now.

We’re thrilled to introduce CData MCP Servers, a new way to securely connect any LLM with Model Context Protocol support to over 350 enterprise data sources instantly, without custom wrappers or risky workarounds. With direct, standardized, SQL-based access, your AI tools can finally talk fluently to your business data.

We're all familiar with wrangling dashboards or multiple systems to get insights into the questions you have at work. At best the answer has been prepared and shared with you in a chart or a report that you just need to find. At worst your request to an overworked operations or business intelligence team goes to the back of their ticket queue.

The power of CData MCP Servers is that you can explore and analyze your business data with simple questions to your LLM. No pre-prepared reports, dashboards, or ticket queues required.

Introducing CData MCP Servers

CData MCP Servers are an implementation of Anthropic’s Model Context Protocol, which was introduced to provide AI tools with secure, two-way connections to external data sources and services. While the MCP standard defines how LLMs should communicate with external systems, it’s the server’s connectivity that determines what those models can actually do.

Each CData MCP Server turns an enterprise application—be it Salesforce, Jira, or NetSuite—into a tabular SQL-based model. This standardized approach allows AI agents to explore the underlying data, discover available entities and attributes, and run queries just as they would against a traditional relational database. Beyond querying, these servers also support write operations and stored procedures, enabling AI-driven workflows and actions.

Our initial beta release includes MCP Servers to 25 enterprise sources and compatibility with Claude Desktop on Windows operating systems. Over the next six weeks, we will rapidly release tranches of additional servers to make our full library of 350+ sources available and expand capabilities with other LLMs and Mac operating systems.

Read more to learn how CData MCP Servers work under the hood.

Follow your curiosity wherever it leads—no ticket required

There’s a lot of buzz around MCP (and for good reason). While other MCP offerings focus on querying data or driving action in a single source-system, CData MCP Servers are ideal for LLM-powered data analysis across multiple data sources. Our CEO and Co-founder, Amit Sharma, explains why CData can uniquely do this in his announcement.

With this initial release, we aim to give everyone their own personal data analyst. We’ve all felt the pain of having a question about your business that can only be answered with the creation of yet another report or dashboard – often relying on data experts or IT to build it for you.

Even then, dashboards can only answer the question they were built for. LLMs with CData MCP Servers adapt to your evolving needs, helping you investigate new angles and test hypotheses on the fly.

Those questions you’ve avoided asking because the answer would require yet another reporting request? Ask them all in one conversation.

Questions like:

• “What if we segmented customers by lifetime value instead of geography?”
• “How do weather patterns affect our delivery times?”
• “Which products perform best with first-time buyers versus repeat customers?”

Built for enterprise AI connectivity from the ground up

We’ve spent over a decade developing one of the largest libraries of high-performance data connectors available and solved for connectivity at enterprise scale. This translates to a big advantage for enterprises looking to provide AI connectivity to their tech stack with:

• Comprehensive data access: We provide a universal SQL layer on top of every source system that’s perfectly suited to large language models. That means your AI tools inherently “speak the language” of your data, eliminating the need for a translation layer or complex API handling. You can ask more natural questions to your AI and it can more effectively translate that into the answers you’re looking for.
• Scalable performance: Our connectors support advanced performance features like query pushdown, parallel paging, streaming mode, and comprehensive bulk operations. Even large-scale queries execute rapidly, ensuring less time spent watching your AI generate responses and more time spent gaining valuable insights.
• Universal toolset across every source: We've created a universal toolset applicable across every supported data source. Whether your LLMs querying a CRM, updating a marketing platform, analyzing financial data, or triggering custom ERP workflows through stored procedures—every action can be taken with the same toolset.
• Respect for user-level permissions and authentication: We access data in-place – never moving it from its source system. This respects the user’s credentials in the source system and only allows data access within their assigned permissions. Whether the source system uses OAuth, SSO (such as Okta or Azure AD), Kerberos, or basic API keys, your AI tool complies with your existing security practices.
• Enterprise-grade security: Additionally, our platform is SOC 2 Type II certified, ISO/IEC 27001:2022 compliant, and GDPR-compliant, ensuring every connection is trustworthy, compliant, and fully audited.

What’s next

We have big plans to build on this initial release:

• Useful prompts, tips, and ideas: Through our limited content series, Vibe Querying with MCP, we’ll offer useful prompts, tips, and ideas to use CData MCP Servers
• Additional sources: We’ll make our full library of 350+ connectors available as MCP Servers
• Expanded compatibility: Expand our offering for full compatibility with Mac operating systems and additional LLM support
• Cloud offering: Package our MCP Servers to be served out of the cloud making it easier and faster for business users to connect their systems
• Support for additional frameworks: Support for additional agent frameworks like the Agent-to-Agent Protocol

Talk to your data – it finally has something to say

The result? Your AI can finally leverage its strengths - insight generation, summarization, and synthesis - drawing accurate, real-time answers directly from the enterprise systems essential to your daily work-life. Plus, data remains securely in place, ensuring your IT team's compliance and security standards are always maintained.

So go ahead - ask Claude to analyze which geographies represent the biggest portion of your revenue. Request ChatGPT to breakdown the biggest product weaknesses driving lost business. Talk with CoPilot about which marketing efforts drive foot traffic to your stores. With CData MCP Servers, your AI is ready to talk to your data—and it finally has something worth saying.

Try it now or Learn more