⌘KCtrl+K
With AI-powered agentic commerce, retailers can enable AI agents to autonomously discover products, negotiate promotions, complete secure checkouts, and deliver personalized post-purchase experiences — all while maintaining full merchant control over pricing, payments, and compliance.
As AI agents become an interface for online shopping, retailers face a fundamental integration challenge. Today's e-commerce infrastructure was built for human-driven browser sessions, not autonomous agent workflows. There is no standardized way for AI agents to discover merchant capabilities, negotiate checkout terms, or process payments securely on a customer's behalf. Without a common protocol, each agent platform requires custom, fragile integrations that don't scale. Meanwhile, merchants risk losing control of critical business logic — dynamic pricing, inventory management, regulatory compliance — when delegating decisions to third-party agents. This gap between agent capabilities and merchant infrastructure limits conversion, erodes margins, and creates friction in the emerging agentic shopping experience.
This NVIDIA AI developer example provides a reference implementation for building secure, protocol-compliant agentic commerce systems that enable AI agents to conduct end-to-end shopping transactions under merchant authority. It shows developers how to implement dual-protocol support — the Agentic Commerce Protocol (ACP) and the Universal Commerce Protocol (UCP) — for standardized agent-to-merchant communication including discovery, checkout, delegated payments, and webhook-driven order lifecycle events. It features NVIDIA AI Enterprise software, including the NVIDIA NeMo Agent Toolkit for multi-agent orchestration, NVIDIA NIM™ microservices for NVIDIA Nemotron LLM for intelligent promotion strategy selection and multilingual post-purchase messaging, NV-EmbedQA-E5 for semantic product embeddings, and an Agentic Retrieval Augmented Generation (ARAG) recommendation pipeline with specialized agents for personalized cross-sell suggestions — delivering an Apps SDK integration with MCP server and merchant-owned widget, Milvus-powered vector search, and visibility into agent reasoning and checkout flow behavior.
NVIDIA NIM Technology
NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal team to ensure this blueprint meets requirements for the relevant industry and use case and addresses unforeseen product misuse.
Please report quality, risk, security vulnerabilities or NVIDIA AI Concerns here.
GOVERNING TERMS: The Blueprint scripts are governed by Apache License, Version 2.0, and enables use of separate open source and proprietary software governed by their respective licenses: Nemotron-Nano-V3, NV-EmbedQA-E5-v5. The sample data is governed by the NVIDIA Data License for Retail Agentic Commerce.
Reference implementation of the Agentic Commerce Protocol (ACP) and Universal Commerce Protocol (UCP) enabling AI-powered checkout negotiation while maintaining merchant of record.