May 21, 2025
3 min read
At PayPal, we continue to make it easier for developers to access our products and services. In early April, PayPal adopted MCP* as the first step to allow developers to embrace the new paradigm of agentic commerce. Today, in support of OpenAI's news, we’re excited to announce PayPal's remote Model Context Protocol (MCP) server now supports OpenAI's large language models (LLMs) — bringing even greater flexibility to developers to accelerate agentic commerce.
The Model Context Protocol (MCP) helps organize and pass the right information to language models to help them work effectively in specific scenarios. PayPal’s MCP server lets developers connect to their preferred AI clients —like OpenAI’s LLMs—through natural language. This setup makes it easy for merchants to handle tasks like creating and listing invoices, generating reports, and accessing valuable insights into merchant transactions.
OpenAI is a natural fit for PayPal’s MCP server integration due to its powerful language understanding, easy-to-use APIs, and support for multilingual, context-aware interactions. It enables merchants to build intuitive, AI-driven workflows that simplify complex tasks like currency conversion, and customer support—all through natural language. With enterprise-grade reliability and flexibility, OpenAI helps developers deliver smarter, faster experiences for users around the world.
With the OpenAI tools that PayPal MCP server provides, developers can build agents for merchants that connect seamlessly to PayPal’s commerce services. Remote MCP server currently offers two valuable tools to developers for merchant integration with more coming soon:
To explore the integration and start building with OpenAI and PayPal’s MCP remote server refer to the developer documentation and watch this demo video.
*Disclaimer: PayPal’s MCP server provides access to AI-generated content that may be inaccurate or incomplete. Users are responsible for independently verifying any information before relying on it. PayPal makes no guarantees regarding output accuracy and is not liable for any decisions, actions, or consequences resulting from its use.
3 min read
3 min read
3 min read