Adding AI chat to a Shopify store without an app
RoutMe has no Shopify app, so the pattern is a script snippet in your theme plus your own read-only tools. Here is what that involves and what it does not.
RoutMe team
Product and engineering
We get asked whether RoutMe has a Shopify app. It does not, and we would rather say so plainly than imply otherwise. There is no Shopify app, no Shopify certification, no partnership and no native connector today. What you can do is simpler than it sounds.
Part one: the widget
Create a chat widget in the RoutMe dashboard and copy the script tag. In Shopify admin, duplicate your theme, choose Edit code, and paste the tag into layout/theme.liquid before the closing body tag. List your allowed origins on the widget exactly, including the www variant and your myshopify.com domain. The step-by-step guide is the Shopify setup guide.
Part two: the data
For the assistant to answer order questions it needs a tool. You run a small MCP server that calls the Shopify Admin API with a custom app token, or the Storefront API for catalog questions, and you register that server in RoutMe. Mark each tool read-only. The options are compared on the Shopify page.
What is coming
Turning a REST or OpenAPI API into MCP tools with one import is in progress, and it would remove most of part two. It is not available yet, and we are not giving a date.
- Shopify
- Widget
- MCP
RoutMe team
Product and engineering
Practical guides from the people who build RoutMe, with every product claim checked against what is live today.
Keep reading
- Product
n8n vs RoutMe for customer conversations: when to use which
n8n is a general workflow engine and RoutMe is built for customer conversations. Here is when each is the right tool, and how teams use both together.
- Engineering
Connect your systems without glue code: MCP, APIs, databases
How RoutMe reaches your data without glue code: MCP servers, your own APIs and read-only database views. What each is, when to use it and how to keep it safe.
See it on your own data
Connect a system, drop an agent node into a flow, and watch it answer from what is actually true in your systems right now.