Prepare product listings and catch catalog inconsistencies
Create a review sheet for missing descriptions, inconsistent variants and unsupported product claims.
What you’ll produce
A proposed catalog-change sheet with source-backed descriptions and a separate list of missing product facts.
Inputs & connections
- A small catalog export with stable product and variant IDs
- Supplier specifications, approved product photos, category rules and brand-writing examples
CSV exports and approved images can support the first pass. A live storefront connection and publication process must be configured separately.
Your starting prompt
Replace the bracketed fields. This is our suggested prompt, not a quote from the source.
Review [catalog batch] using [approved specifications, photos and category rules]. Preserve every product and variant ID. Flag duplicate-looking records, missing fields, inconsistent units and category mismatches. Draft concise descriptions using only supported attributes; do not infer materials, compatibility, certifications or included accessories from a photo. Keep suggested edits separate from unchanged fields. For every edit provide old value, proposed value, supporting source and reason. Use unknown when a fact is missing. Return an import-ready draft plus questions; do not change prices, remove branding or publish listings.
How to run it
- 01
Export a small product batch and save the unmodified file.
- 02
Specify required fields, category rules and facts that need a supplier source.
- 03
Generate the change sheet and check variant relationships and product IDs.
- 04
Have a merchandiser verify claims and approve individual changes.
- 05
Test the approved import on a small batch and inspect the storefront before expanding.
Review before you act
Check variant mappings, measurements, product claims, accessibility text and whether descriptions match the actual item.
Limitations & success checks
- A product image rarely establishes exact material, size or compatibility.
- Storefront imports may overwrite fields or variant relationships if identifiers are wrong.
Measure: Accepted edits, unsupported claims removed, variant errors and missing fields resolved.
Sources & attribution
The sources below support the reported example or a tool capability. They do not validate our prompt or prove the outcome in your environment.
Source post and author checked through X public oEmbed on October 17, 2026. The post describes Shopify merchant tasks with Claude and also promotes Shoptank and Libautech products. Treat it as a promotional practitioner example, not a neutral comparison or proof of ranking gains. The export-and-review process, prompt and checks here are original editorial additions; no built-in Shopify connection is promised.
Creator / publisher: @DavisPaipa
Editorial review: 2026-10-17. How we label evidence.