Add product schema so AI shopping assistants can read your catalog

AI shopping assistants and search features increasingly read a page’s structured data rather than guessing from the layout. If your product pages ship clean Product schema, with a price, availability, and SKU an assistant can parse, your catalog is legible to them. If they do not, an assistant is left scraping and often gets it wrong.

This builds that schema from each product’s own fields and writes it through Rank Math. It reads first, shows you the schema, and only writes what you approve. One honest limit before you start: this adds schema on the page, it does not build a product feed.

I want to add Product structured data to [the products in the "Accessories"
category] so AI assistants and search can read them. This store uses Rank Math.

First pass, write nothing: read each product's name, description, price, SKU,
and stock status, then build the Product schema object you would attach to it
and show me the schema for the first [5]. Use only the values already on the
product, do not invent anything. Let me check them.

Only after I approve, write the schema through Rank Math, one product at a
time in small batches, and show me each batch before it saves. At the end,
list every product you added schema to.

Why this is safe to run

The first pass is reads. Reading each product’s fields writes nothing, and the schema object is shown to you built entirely from values already on the product, so you can see it is accurate before anything is stored.

The honest scope matters here. This writes Product schema onto the product page through Rank Math, which is exactly what an on-page assistant parses. It is not a product feed: the file a shopping engine like Google Merchant Center ingests is generated by a dedicated feed plugin, and building that is outside what these abilities do. The write reaches one ability, updating a post’s Rank Math schema, and it stays off until you enable it. Schema is stored in the product’s own Rank Math meta, so a wrong entry is re-writable through the same flow. Work runs in small batches you approve, each call is checked against the capability of the user you bound the agent to, and every write lands in the Activity Log with the product it touched.

Already connected? You do not have to paste this at all. Ask your agent for the “add product schema for AI shopping assistants” recipe, name the set, and it runs the same read, approve, then write flow on your own store.