Making e-commerce agent-ready with Worldline

AI agents are changing the way we organise our lives, work, search, and shop online and how businesses operate. From copilots embedded in daily workflows to autonomous agents that can plan and act, large language models (LLMs) are becoming a new interface for software and services.

The next big evolution in e-commerce and payments is infrastructure that is easily accessible and interoperable with AI agents. To make AI truly useful and efficient in e-commerce, it needs a secure and standardised way to connect to real-world systems, tools and data - such as product catalogues, order management tools, and payment systems, to power the end-to-end user journey. The Model Context Protocol (MCP) provides that missing link.

Today, we’re launching the Worldline MCP Server, making Worldline’s payment capabilities accessible to AI agents in a secure and developer-friendly way.

Introducing Worldline MCP Server

Worldline MCP server bridges our payment functionalities with AI agents. With the MCP server, we provide a new way to access and integrate with our payment platforms. It acts as a translation layer between LLMs and our payment APIs, allowing AI agents to trigger payment actions using natural language prompts instead of complex integrations.

For merchants, MCP Server opens new avenues for innovation. Merchants can experiment with AI-powered commerce use cases and build intelligent payment workflows with AI. With our MCP Server, developers can:

• Access Worldline payment functionality through LLMs
• Build AI-powered commerce workflows faster
• Experiment with agent-based operations and payments without rewriting existing code

We are launching a Commerce MCP Server and an Operations MCP Server tailored for different use cases. For example, a merchant building its own AI shopping agent can use the Worldline Commerce MCP Server to share a secure payment link with a consumer to complete a transaction. A developer can instruct an AI tool such as Gemini Code Assist to “Create a refund for transaction number XXXXXX” with the Worldline Operations MCP Server. In these scenarios, the MCP Server interprets the request and executes the corresponding payment action. Other use cases include getting available payment methods for consumers to pay for an agent-initiated transactions, retrieve payment status, or capture funds — helping merchants experiment and benefit from new payment use cases and AI-assisted workflows.

The Worldline MCP Server supports a growing set of Worldline Connect payment APIs, you can learn more and start experimenting via our Documentation Site.

Our MCP Server simplifies how AI agents communicate with Worldline's payment services, ensuring secure and efficient transaction processing.

Agentic Commerce: building for what’s next

MCP Server is our first building blocks for Agentic Commerce. We are also working on preparing merchants for Agentic Commerce, the next evolution of online shopping, where autonomous AI agents act on behalf of consumers or businesses to search, evaluate, decide, and transact.

The ecosystem is already moving in this direction. Major players such as Visa, Mastercard, Google, and OpenAI are introducing new protocols and frameworks designed to support trusted, AI-initiated transactions. While standards and use cases are still evolving, it is clear that merchants will need infrastructure that allows them to interact with AI agents securely, compliantly, and at scale.

As part of this launch, Worldline is introducing ConnectAI, a dedicated section on our developer portal focused on building with AI. ConnectAI brings together our work on:

• Agentic commerce offerings
• Emerging agentic payment protocols, including Google’s Agent Payments Protocol (AP2), OpenAI’s Agentic Commerce Protocol (ACP), and Visa and Mastercard frameworks

Through ConnectAI, developers and merchants can explore how to build, test, and prepare for agentic commerce — starting with MCP-enabled payment workflows today and evolving toward agent-initiated payments with Worldline.

If you are interested in experimenting with AI-powered commerce or piloting agentic commerce use cases with Worldline, we’d love to hear from you.

Quick links:

• Worldline Documentation Site
• Worldline ConnectAI