Loading...
Conor Redmond (Director & Product Lead for AI), Colm Reid (Director, API Design & Delivery), Darshit Metaliya (Lead Software Engineer & Architect), & Mark Turley (AI & Machine Learning Engineer)
The world of e-commerce and software development is on the cusp of a major transformation. The traditional methods for building and maintaining online businesses are being challenged by a new wave of AI-driven development and agentic commerce. We're thrilled to announce a significant leap forward in this new era: the launch of our Model Context Protocol (MCP) server.
We have developed a locally hosted MCP server that allows developers to interact with the Global Payments API using natural language. This NPM package can be easily installed on a local machine. It functions similarly to an SDK or a library.
This initial version includes two core tools:
In addition to this, we will soon be adding the remaining endpoints to wrap the entire API, including the Actions Endpoint, which enables you to view logs from your recent interactions with our API.
Our MCP server is already supported by several popular AI platforms, including Claude Desktop, VS Code, and Windsurf.
The modern developer workflow has evolved. Tedious searches through documentation and complex integrations are being replaced by AI-powered assistants. Our MCP server bridges the gap to make our powerful API more accessible than ever. These advancements are not just about keeping up with the times; they are about leading the charge into the future of agentic commerce.
Agentic Commerce, where AI agents perform research and execute financial transactions on behalf of users, is projected to become a multi-trillion dollar market. MCP is a core part of delivering on the promise of Agentic Commerce.
By providing the foundational plumbing for this new ecosystem, we're empowering the next generation of developers and entrepreneurs.
The MCP server represents more than just a technological advancement, or a new concept. It embodies a more startup-like approach to innovation. Recognizing the rapid pace of change within the AI landscape, we intentionally deviated from our traditional development lifecycle. Instead, we adopted a "hackathon-style" mentality, giving our team the autonomy to make a strategic bet on the future without getting bogged down with the need for a perfect, predetermined business case. Our philosophy was simple: in a world of uncertainty, it's better to launch quickly, learn from real-world feedback, and iterate. This project served as a powerful proof of concept for this faster, more dynamic way of working.
The possibilities created by MCP are vast and exciting. Here are just a few examples of what's now possible:
This is just the beginning. We have a clear roadmap for the future of MCP at Global Payments:
Expanded API Coverage: As mentioned, earlier, we plan to wrap all of our API endpoints into the MCP server very soon, and we expect to make quick progress in this area. Keep an eye on our github for the latest updates.
Streamable MCP: We are already working on a streamable MCP that we will host as a dedicated endpoint, making integration even easier.
A Collection of MCPs: We envision a future with a collection of MCPs that can be combined through an orchestrator to provide a single, unified view of the world.
The launch of our MCP server is a testament to our commitment to innovation and our belief in the power of AI to transform the payments landscape. We're excited to see what you'll build with it