Direct answer: Machine Interface (MI) is the set of access surfaces through which an AI agent can retrieve Shopify commerce facts and, where permitted, request an action. It is a supporting capability within Machine Experience, not a fourth public framework.
This article defines MI. For its quality boundary, read MI versus MX; for merchant-specific testing, use the MI audit. An interface's existence does not prove that the information it returns is accurate.
AI shopping agents cannot work directly from merchant intent. They need an interface through which products, variants, prices, inventory, delivery information, policies and permitted actions become accessible. Codified Commerce calls this layer the Machine Interface, or MI.
Machine Interface definition
A Machine Interface is the collection of machine-accessible surfaces, structures, protocols and permissions through which an AI system can retrieve commerce information and perform authorized actions.
It may include structured page data, Shopify APIs, product feeds, app tools, webhooks, search endpoints, cart operations and checkout handoffs. The interface is not a single technology. It is the boundary between a commerce system and the machines attempting to use it.
Why Shopify merchants need an MI layer
A conventional storefront is designed primarily for people. A shopper can interpret visual hierarchy, tolerate missing context and ask a support team for clarification. A machine needs explicit entities, relationships, current values and predictable operations.
MI is not the same as an API
An API is one possible interface. A merchant’s broader MI can include publicly retrievable content, schema markup, catalogue feeds, Shopify storefront capabilities and approved tools.
Information and action
A useful commerce MI supports read interfaces for discovery and evaluation and action interfaces for authorized operations such as variant selection, cart creation and checkout handoff.
Agentic Payments: where MI becomes transactional
Agentic Payments makes the Machine Interface commercially consequential. Once an authorized agent moves beyond discovery toward checkout, the interface must expose correct products, variants, price, availability, policies and transaction capabilities. UCP is one example of a structured commerce interface, while the Global Agentic Payments Readiness Framework shows how MI connects to checkout, payment compatibility, identity and governance.
Where MI fits in the Codified Commerce system
Machine Interface exposes information and capabilities. Machine Experience evaluates whether machines can understand and use them reliably. Machine Learning interprets interaction and outcome signals. Moving Target detects meaningful change, while Next Product Move determines what the merchant should consider changing.
Continue with Machine Experience, Machine Learning for Shopify, Moving Target, and Next Product Move.
Explore the Machine Interface cluster
Explore Machine Experience →
Discuss a controlled Shopify automation →
