Think inside your AI world.
How do I connect Shopify to Claude?
Shopify's MCP story is plural: official Storefront, Customer-Accounts and Order servers, with community admin servers alongside — pick the surface that matches the question.
Direct answer: Shopify ships official MCP servers for the storefront (catalogue, product and variant detail, cart, policies), customer accounts and orders (status, details, preferences), with community admin servers extending to products, orders, customers and inventory. In Claude: Settings → Connectors → add the server for the surface you need and authorise; Claude Code takes 'claude mcp add'.
Which surface answers which question
Catalogue and variant questions read from the storefront surface; order status and customer detail from the orders surface; inventory and admin operations from an admin server. Matching surface to question keeps grants minimal — a support workflow needs orders, not admin.
The reorder maths the store doesn't do
Worked case: you ratified a stockout rule — reorder when stock cover drops under the supplier's lead time plus a fortnight's buffer, priced by a bestseller that went dark for three weeks in Q4. Shopify tracks stock levels exactly; lead-time cover is arithmetic between your supplier reality and your risk appetite, and neither lives in the store.
Inventory, read against cover
Through Unl the stock read arrives computed and judged: two SKUs under your cover line — reorder now, by the rule Q4 wrote. The weekly stock scan becomes a verdict list. Same store data; your operational arithmetic finally standing over it.
Shopify to Claude reads the store surface you grant; Shopify through Unl reads it against the operating rules you ratified — cover lines, margin floors — so stock questions come back as decisions.
Reads through Unl arrive with measured context — in the presence of the decisions you’ve already settled. The reach lane is live: one box, paste anything. If it speaks MCP, Unl can reach it. Readings arrive unprompted, the data beside the criterion; Unl is a courier, not a warehouse, and keeps only your keys and the frame.
Read further
Questions people ask
Which Shopify MCP servers are official?
Storefront (catalogue, product/variant detail, cart, policies), Customer Accounts and Order servers are Shopify's own; community admin servers — StackOne's 85-action server among them — cover products, orders, customers and inventory.
Can ChatGPT connect to the same servers?
Yes — every server named here speaks standard MCP; 'shopify mcp' resolves the same way from any capable client.
What store rules pair well with the connection?
Cover-based reorder points, margin floors per SKU, a discount-depth ceiling. Each is a number plus a reason — the pairing that turns a stock or pricing read into an answer.
What if a promotion deliberately goes under the margin floor?
Mark it as the exception it is. A floor with recorded exceptions still tells you something — you can see how often you go under and why. A floor that is quietly ignored during every promotion has stopped being a floor and nobody has said so.
What this is
Think inside your AI world — you stay in command
Save the thoughts, decisions and targets worth keeping, each with its reasoning, carried into every AI session the moment they matter. A new unit of exchange between you and your AI: the Settled Why with standing that travels. Unprompted.
Your whole AI world. What you decided at the epicentre. It plugs into Claude, Claude Code, ChatGPT and Cursor as an MCP connector — quick to connect, in a couple of steps.
MCP native·Human settled·Model agnostic·Your data
Measured Context
Connect Unl to bring the right information into the moment.
Your sources, read against the criteria you set.
The full product, open. Free at launch.