All use cases

Write descriptions and SEO for product cards

Descriptions were copied from the supplier and the listing never shows up in search. The agent rebuilds the title, attributes and copy around real queries.

Medium · 30 min · on an event

What you get

SKU 10457 — Ozon

Title: Thermal mug 450 ml, stainless steel, black
Queries placed in fields: thermal mug 450 ml, steel travel mug
Body: 6 of 6 attributes taken from your spec sheet

Left empty: country of origin, warranty. The agent leaves
these blank rather than inventing them.

A sample on made-up data — your numbers will be your own.

Who it fits

  • Hundreds of listings, descriptions copied from the supplier, invisible in the platform's search.
  • A product is going onto a second platform and the text needs rebuilding for its fields and limits.

When it won't work

The product's real specifications are empty: blank fields produce text without a single fact.

How the agent does it

1

Pick listings and data

Give the article numbers and everything you have: spec sheet, measurements, composition. More facts means fewer generic words in the copy.

2

Run the rework

Queries are spread across the fields instead of dumped at the end. Anything absent from your source data gets flagged.

3

Check against the item

A description is the seller's statement about the product. One invented line costs more than a search position.

What you set

Connect Ozon and Wildberries. Manually: the list of listings, the supplier spec sheet, and priority queries if you already have them.

What you'll need

Ozon
Wildberries

Starter prompt

Copy the prompt or open it straight in a chat with the agent.

Prompt for the agent

Rewrite the listings [article numbers] on [Ozon / Wildberries]: title, specifications and description. Take buyer queries for this category and distribute them across the fields: the main ones — into the title and the first specifications, the rest — into the description, without listing them comma-separated. Write only from the specifications in the listing and from what I attached. Don't add composition, country of origin, warranty, certificates or product properties: if a field is empty, put [to clarify] and move it into the list of missing data. Respect the platform's field limits and don't repeat a single keyword more than [3] times. Output for each article number: the title, a specifications table, the description text, the list of queries used and the list of fields I have to fill in myself.

Open in chat
Check the result

Check the listing descriptions. List: — specifications that weren't in the source data — composition, country, warranty, certification; — claims the platform might consider unprovable: "the best", "hypoallergenic", "certified"; — fields where the character limit is exceeded or the platform's structure is broken; — queries you inserted for search volume rather than for the product's meaning; — listings where the description came out generic because there weren't enough facts. At the end: which listings can be uploaded and which are waiting on data from me.

A second prompt — the agent uses it to review its own work and show what's left for you.

Related use cases