Commerce and Cloudflare make BigCommerce stores agent-ready by default

WebMCP lets AI shopping agents query live inventory and pricing directly, bypassing the scraping era of agentic commerce.

By Disrupts Editorial Team Sat, 7 November 2026 AI level 2 of 5

AI level 2 of 5: drafted by our AI editorial assistant from the company's announcement; fact-checked, edited and signed off by Mark Walker, Editorial Director. What the levels mean

A long data center aisle is lined with dark server racks displaying glowing blue and green indicator lights, illuminated by bright overhead lighting.

Commerce.com (Nasdaq: CMRC), the parent company of BigCommerce, has partnered with network infrastructure giant Cloudflare to deploy WebMCP across eligible BigCommerce merchant storefronts by default. The integration, announced on 21 October 2026, means that AI agents operating inside a shopper's browser can query live product catalogues, check real-time stock levels, and initiate checkout flows without merchants writing a single line of code or altering their existing site architecture.

The move is a direct response to a structural shift in how consumers discover and purchase products online. As AI assistants become embedded in browsers and operating systems, the legacy assumption that storefronts need only be legible to human eyes is being dismantled at speed. Most existing e-commerce infrastructure was designed for point-and-click navigation; AI agents attempting to shop through those same interfaces have historically resorted to scraping page content and guessing at form fields, producing inaccurate prices and incorrect inventory data. WebMCP replaces that improvised behaviour with a formal protocol handshake between the shopper's agent and the merchant's store.

From scraping to structured dialogue

WebMCP builds on BigCommerce's existing MCP Toolkit, which exposes a set of shopping actions that agents can call to complete real-world tasks. The server-side Storefront MCPs handle agents connecting from outside the storefront, while WebMCP extends the same structured access to agents working directly inside a browser session. The practical outcome: when a shopper instructs a browser agent to find, say, a lightweight jacket under £120 in a specific size, the agent can interrogate the store's actual catalogue, surface the closest match, add it to a cart, and return a checkout link, all without the merchant configuring anything.

"WebMCP replaces guessing with a handshake," said Sharon Gee, senior vice president of product for AI at Commerce. "A shopper's agent asks the store directly for what's in stock at what price in which sizes, and the store answers accurately."

Critically, the merchant retains ownership of the checkout flow, customer relationship, and brand identity throughout. Commerce has also introduced a centralised settings menu that governs WebMCP and all other AI connections to a merchant's store, giving operators the option to disable the feature entirely. That control layer matters: as agentic purchasing scales, questions around liability for agent-initiated transactions, data residency, and consumer protection are likely to attract regulatory attention on both sides of the Atlantic.

The broader agentic commerce landscape

The Commerce-Cloudflare announcement is one of the clearest signals yet that agentic commerce is moving from experimental pilots into default infrastructure. Shopify has explored similar agent-readiness tooling, and Google's Shopping Graph already feeds structured product data to AI surfaces including Gemini. What distinguishes the WebMCP approach is its reliance on an open standard rather than a proprietary feed, positioning it closer to the emerging Model Context Protocol (MCP) ecosystem being adopted across a widening range of enterprise software.

For investors and cross-sector strategists, the implications extend well beyond retail. The same protocol logic that lets an AI agent query a jacket's availability could govern agentic procurement in B2B supply chains, pharmaceutical catalogue access, or even energy trading platforms where automated agents need verified, structured data rather than scraped estimates. Commerce explicitly notes that WebMCP applies to both B2C and B2B storefronts, hinting at the enterprise procurement angle that could substantially expand the addressable market.

The capital landscape around agentic infrastructure is intensifying. Cloudflare's willingness to co-develop and co-announce at this level of integration reflects a broader strategic bet that the network layer becomes the trust and routing layer for agent-to-service communication. For platform-layer players like Commerce, the race is to become the default agent-ready commerce substrate before that role is claimed by the hyperscalers or the AI labs themselves. Merchants who adopt early inherit a structural discovery advantage; those who do not risk invisibility as AI agents increasingly bypass storefronts that cannot respond to structured queries.