Showcases a complete sandbox implementation of OpenAI's Agentic Commerce Protocol, demonstrating agent-driven transactions between an LLM client, merchant, and payment service provider.
Sponsored
Explore the foundational Agentic Commerce Protocol (ACP) released by OpenAI with this first working mock implementation. This open-source sandbox provides an end-to-end demonstration of how LLM clients can seamlessly transact with vendors and payment providers, replicating the ACP flow with distinct Client, Merchant, and Payment Service Provider (PSP) components. It's designed to accelerate experimentation and understanding for developers building the next generation of agentic commerce applications, supporting both OpenAI and Anthropic models.
Key Features
01Reference implementation of OpenAI's Agentic Commerce Protocol
02End-to-end demo of Client, Merchant, and PSP interaction
03MCP-UI compatible chat interface for agent interaction
04Supports both OpenAI and Anthropic LLM API keys
05Implements Delegated Checkout and Product Feed specifications
062 GitHub stars
Use Cases
01Experimenting with the Agentic Commerce Protocol (ACP)
02Developing and testing agentic e-commerce applications
03Understanding the technical flow of agent-driven payments and transactions