The industry is quickly moving from isolated, static channels to autonomous, context-aware agents that coordinate purchasing experiences smoothly across stores, digital platforms, and conversational interfaces.

Agentic AI executes complex workflows, manages supply chains, and optimizes pricing in real-time. By moving away from isolated “bots” and toward a networked architecture of agents, retailers can finally achieve the agility required for the modern market.

Agentic commerce represents a transformative shift toward a retail model where humans and AI agents operate as a cohesive team, unified by shared context and open protocols.

This ecosystem is powered by the Model Context Protocol (MCP), which serves as a centralized “brain” providing agents with an enterprise-grade understanding of inventory, pricing, and customer intent. By grounding agents in this common business logic and leveraging the Agent Communication Protocol (ACP) to bridge functional silos, retailers can ensure that agents in merchandising, supply chain, and store operations collaborate in real time.

The practical application of this model is realized through the Dynamics 365 Commerce MCP Server, which is expected to enter preview in February 2026. By exposing core functions—such as catalog management, promotions, and fulfillment—as MCP-enabled capabilities, the platform connects back-office ERP and analytics directly to front-office experiences. This integration enables retailers to build agile, agent-driven workflows in which AI can securely discover and execute tasks across the entire value chain.

Additionally, this protocol-driven approach empowers “Commerce Anywhere,” providing the speed required to meet the demands of modern, multi-channel retail.

There are three ways in which you can find an agent: partner-made one, build a custom one with MCP, or use the existing standard Dynamics 365.

Retailers can begin by using agents already built into Dynamics 365.

Because each retailer has its own set of rules and demands, it can utilize Microsoft Copilot Studio to create customized agents. These agents can be tailored to adhere to specific merchandising strategies or replenishment guidelines. These custom agents leverage MCP-powered access to enterprise systems, enabling seamless integration into existing workflows.

On the customer-facing side, the Dynamics 365 Commerce MCP Server extends this intelligence directly into the shopping experience. Custom agents can discover products, personalize offers, and even complete transactions across digital and physical channels.

Examples of agents in the industry:

The transition to agentic commerce is more than a technological upgrade; it is a strategic imperative for retailers who want to thrive in a decentralized, fast-paced market. By adopting a protocol-based approach through MCP and ACP, businesses can finally break down the walls between the back office and the front line. Whether you start by deploying embedded agents for supply chain monitoring or collaborate with partners to create bespoke clienteling experiences, the goal remains the same: to create a seamless, intelligent, and highly responsive retail environment.