In the rush to adopt Artificial Intelligence, most enterprises face a common roadblock: How do we make our existing, complex business logic accessible to AI agents without rebuilding everything from scratch? At TIBCO, we believe the answer isn’t a “rip and replace” strategy. It’s about interoperability. That is why we are excited to announce a strategic update to TIBCO BusinessWorks™ – Native Model Context Protocol (MCP) Server Support for REST and SOAP Services.
What is Model Context Protocol and Why Does It Matter?
The Model Context Protocol (MCP) is a standardized open protocol that enables AI agents to seamlessly discover and interact with data and services. Think of it as a universal translator.
Historically, connecting an LLM (Large Language Model) to a backend SOAP or REST service required custom “glue code” or manual prompt engineering. By natively supporting MCP, TIBCO BusinessWorks allows your AI agents to “see” and “use” your enterprise functions as if they were native tools.
The Strategic Shift: TIBCO as the AI Backbone
This update represents a core shift in TIBCO’s AI strategy. We aren’t just adding AI features to our products; we are making our products the reliable infrastructure that fuels Enterprise AI.
By exposing BusinessWorks services as MCP servers, we are solving three critical challenges:
- Dynamic Discovery: AI agents can now query your TIBCO environment in real time to determine which functions are available.
- Legacy Modernization: You can bring decades of proven SOAP and REST logic into the AI era without a single line of code change to the original service.
- Standardization: Using a standardized protocol ensures that your architecture remains vendor-agnostic and ready for whichever AI model you choose next.
Seamless Implementation: From API to AI Tool
The transition from a standard service to an AI-ready tool is designed to be frictionless for developers. Within TIBCO Business Studio™, enabling this capability is a matter of configuration, not reconstruction:
- One-Click Tool Exposure: Developers can simply right-click an existing service operation and select “Use as a Tool” to expose it to the MCP server.
- Built-in Discovery: The new MCP Server View enables management and naming of these tools, providing the descriptions and metadata that AI agents need to understand when and how to use them.
- Command-Line Automation: New bwdesign utility commands, allow teams to automate the AI enablement of existing projects or spin up new MCP-ready skeletons from API specs.
Enterprise-Grade Security for AI
We understand that “opening up” services to AI agents raises significant security concerns. TIBCO’s MCP implementation is built with enterprise rigors at its core:
- Authentication: Supports Basic Authentication and JWT (JSON Web Tokens).
- Granular Authorization: You can restrict tool access using specific JWT scopes (e.g., read: tools or update: tools), ensuring an agent can only perform the actions it is authorized to do.
- Secure Communication: Full support for SSL configuration, ensuring all interactions between the AI client and the TIBCO server are encrypted.
Ready to build and experiment locally? Simply enable the MCP Server View in TIBCO BusinessStudio and start exposing your APIs as MCP tools. It is available in Tech Preview with TIBCO Businessworks 6.12.0 HF02.
The Bottom Line: Real-World Value
For the enterprise, this update significantly reduces the time-to-market for AI initiatives. Instead of rebuilding months of business logic—such as inventory checks or customer lookups—teams can now expose entire libraries of proven SOAP and REST logic to AI agents in minutes. This provides the robust, governed, and scalable “plumbing” required to move AI from experimental chat interfaces into production-ready business automation.
Author:
Mustafa Tavawala
Mustafa Tavawala is a Lead Product Manager at TIBCO and is primarily responsible for designing and executing the roadmap for the TIBCO Businessworks 6 product suite. He has also worked as a Solution Architect with hands-on experience leading application and data integration initiatives for a diverse set of customers.
