thinkhard.SI

Prompt repository Small business

Product description prompt: a listing from your own specs

You sell on Etsy, Shopify, Amazon or your own site and need a title, bullets and description built from the facts you have, not invented features.

Registry no.
SMB-002
Collection
Small business
Laid out for
GPT
Score
10 out of 10
Blanks to fill
7
Added

Small businessGPT

The prompt

Score 10 of 10, Excellent

Use it when You sell on Etsy, Shopify, Amazon or your own site and need a title, bullets and description built from the facts you have, not invented features.

  • {{marketplace}}
  • {{target_buyer}}
  • {{title_limit}}
  • {{tag_count}}
  • {{product_specs}}
  • {{keywords}}
  • {{shop_voice}}
Full prompt and why it works
Prompt
# Role
You write product listings for a small shop that sells on {{marketplace}}.

# Goal
Write a complete listing for the product in the Context section, aimed at {{target_buyer}}, so a shopper can decide to buy without having to message us with questions.

# Success criteria
- Every feature, material, size, care instruction and compatibility claim comes from the specs. A shopper who receives the product finds it exactly as described.
- The first sentence of the description says what the product is and who it is for. The rest answers what buyers ask before buying, such as size, materials, what is included, and care.
- Search terms from the keyword list appear where they read naturally.

# Constraints
- Do not invent dimensions, materials, certifications, health or safety claims, labels such as "handmade" or "eco-friendly", or comparisons with other brands.
- No superlatives such as "best" or "perfect", and no urgency language such as "selling fast".
- Title: under {{title_limit}} characters, with the most important search terms first, readable as a phrase rather than a list of keywords.

# Output
1. Title, with its character count.
2. Five bullet points, each starting with a short bold label, under 200 characters each.
3. Description: 120 to 220 words in short paragraphs, ending with the care and shipping notes from the specs.
4. Up to {{tag_count}} search tags, comma separated.
5. Questions a buyer may still have: anything the specs do not answer.

# Stop rules
If a spec that buyers of this kind of product usually need is missing, such as size or material, do not fill it in. Leave it out and say so under the last heading.

# Context
<specs>
{{product_specs}}
</specs>

<keywords>
{{keywords}}
</keywords>

<shop_voice>
{{shop_voice}}
</shop_voice>

Variables

{{marketplace}}
Where you sell, for example Etsy or your own Shopify store.
{{target_buyer}}
Who buys it, for example "people furnishing a first apartment".
{{title_limit}}
The platform's title limit, for example 140. Check its current rules.
{{tag_count}}
How many tags the platform allows, for example 13.
{{product_specs}}
Everything true about the product: what it is, materials, size, colors, what is included, care, shipping, who makes it.
{{keywords}}
Search terms buyers use, comma separated.
{{shop_voice}}
A few words on how your shop sounds, or a past listing you like.

Why it works

  • What to avoid. It rules out invented measurements, materials and labels like "eco-friendly", which lead to returns, bad reviews and marketplace policy trouble.
  • What good looks like. A shopper who receives the product should find it exactly as described; that test keeps the copy honest.
  • If unsure. Specs buyers usually need but you did not give are listed as open questions instead of filled in.

How to use it

  1. Press Fill in variables and type your details into each blank. What you type stays in this browser.
  2. Copy the filled prompt into the AI tool you already use, such as ChatGPT, Claude or Gemini, or open it in the Grader or the Prompt Tester here.
  3. Before you rely on it, try it on three or four real examples. The Prompt Tester runs them all at once and shows which ones go wrong.
Use it from code

Get this prompt as JSON from the free REST API, with your own text in the blanks, or let an AI assistant fetch it through the MCP server. No key.

curl -s https://thinkhard.si/api/v1/prompts/product-listing-from-specs

With the MCP server added to Claude, ChatGPT, Cursor or VS Code, call get_prompt with {"prompt": "product-listing-from-specs"}, or just ask for it by name.