Skip to content

Used part pricing from photos

Reads part numbers from a folder of photos, looks up live marketplace prices and builds a priced spreadsheet ready to list. You check the first row before it does the rest.

Ecommerce and resaleWorks with Claude Code · OpenAI Agents SDK+1Community example

What it does

  • Reads the part number and a short description from each photo
  • Searches marketplace listings for each number
  • Picks a consensus price and rounds to your price rule
  • Writes one spreadsheet row per part, with failed ones marked

Your approval boundaries

Approval required

These actions always wait for your yes:

  • Approving the first finished row before the full run
  • Listing any part for sale
  • Running past the spend cap

You decide what happens next.

More about this workflow

This setup is for online sellers and resale shops juggling listings, orders and buyers. It takes a repeat chore off your plate and hands back something you can check quickly.

First, it reads the part number and a short description from each photo. Next, it searches marketplace listings for each number. Then it picks a consensus price and rounds to your price rule. Finally, it writes one spreadsheet row per part, with failed ones marked. It works with your photo storage, marketplace api and spreadsheets, each on its own login with limited access.

You get the prep work done and waiting for you. The agent stops and asks for your yes before approving the first finished row before the full run, listing any part for sale and running past the spend cap, and says plainly if a step cannot be undone.

“before the full run you approve the first finished line”
tamaugo tamaugo on GitHub · Sep 25, 2026
Before you startPhoto storage · Marketplace API accessScoped permissionsClaude Code or OpenAI Agents SDK
Explore required tools and model recommendations

Tools and access

  • Photo storage

    Handles the photo storage part of this job. Grant read-only access to one photo folder.

    API required: No. A shared folder or drive for item photos. No API needed if the agent can read it.

    No sign-up needed
  • eBay

    Handles the marketplace api part of this job. Grant read scopes for orders and listings.

    API required: Yes. Needs an eBay developer account and an app keyset. Seller actions use OAuth tokens with the scopes you grant.

  • Google Sheets

    Handles the spreadsheets part of this job. Grant access to the single sheet it needs.

    API required: Yes. Needs a Google Cloud OAuth app or a service account with the Sheets API turned on.

Official links only. Not sponsored, no affiliate links.

Model recommendations

Best pick

Gemini 3.8 Flash

Built for agent workflows, with a free tier to test on.

$0.75 per million input tokens and $3.75 per million output tokens through December 31, 2026. $1.50 and $7.50 from January 1, 2027.

Google docs as of October 11, 2026

Cheaper backup

Gemini 3.5 Flash-Lite

Low cost for many listings at once.

$0.30 per million input tokens, $2.50 per million output tokens.

Google docs as of October 11, 2026

Our own picks. Not sponsored, no affiliate links.

Build your prompt

Add your details, then copy the prompt into your tool.

1Your details

Fill these in and the prompt updates as you type.

make, model, year

floor and rounding

2Your prompt

4 of 4 still empty
Fill these in before you run it: [THE PHOTO FOLDER] [THE VEHICLE OR PRODUCT LINE: make, model, year] [YOUR PRICE RULE: floor and rounding] [SPEND CAP PER RUN] You are the Used part pricing from photos agent. Reads part numbers from a folder of photos, looks up live marketplace prices and builds a priced spreadsheet ready to list. You check the first row before it does the rest. Use only the details filled in above. If any [BRACKETED] item is still blank or unclear, stop and ask for it first. Do not guess. What you do on your own: - Reads the part number and a short description from each photo - Searches marketplace listings for each number - Picks a consensus price and rounds to your price rule - Writes one spreadsheet row per part, with failed ones marked What you must never do without a yes from a person: - Approving the first finished row before the full run - Listing any part for sale - Running past the spend cap Rules: - Before any step on the list above, stop and ask. Say in plain words what you want to do, who it affects, and whether it can be undone. - If something cannot be undone, say so first, in one short line. - Write drafts, not final actions. A person sends, pays, posts, or deletes. - Keep a short log of every action you took and every source you read. - If you are unsure, ask instead of guessing.
View run instructionsPaste into: Claude Code, OpenAI Agents SDK, Make

Your answers stay in this browser only.

Run your workflow

Connect your tools with only the access this setup needs, then follow these steps.

Steps

Do these in order. Test on old data before it touches live work.

  1. 1Set up Photo storage. Keep it to read-only access to one photo folder.
  2. 2Connect eBay with its own login, not yours. Grant read scopes for orders and listings.
  3. 3Connect Google Sheets with its own login, not yours. Grant access to the single sheet it needs.
  4. 4Paste the prompt where your harness takes instructions. See the note under the Copy button.
  5. 5Do a dry run on old or test data first. Check every draft and every stop point by hand before you let it near live work.
  6. 6Read its log after the first real run. Only widen access once a week of runs looks clean.

Safeguards and trust score

The score describes this example setup’s access, approval boundaries, ability to undo changes, and scope.

Access18 / 30

Can draft things that send or spend, so it sits close to risky actions.

Human approval30 / 30

Waits for a yes on: approving the first finished row before the full run, listing any part for sale, running past the spend cap.

Undo and logs14 / 20

Proposes changes to live records, so the setup needs a change log and a way back.

Scope16 / 20

Runs on its own login across 3 tools. Fewer connections means a smaller blast radius.

How to raise it

  • Access: Take away its ability to send, post or spend. Let it only write drafts, and have a person press send or pay from their own account +6
  • Undo and logs: Stop it from editing live records. Have it write the change as a proposal you apply, and keep a log of every change with a way to roll it back +3
  • Scope: Cut it down from 3 connected tools to the 2 it needs most, each with its own limited login instead of your full account +2

Do all of these to reach 89, up from 78.

Want help getting there? Get an AI Trust Audit from Pixeldesigns

Reviewed by Trustcase (example setup). How scoring works

Embed badge

Show this trust score on your site or README. It links back here.

Trustcase trust score badge
Markdown
[![Trustcase trust score for Used part pricing from photos](https://trustcase.dev/badge/used-part-pricing-from-photos.svg)](https://trustcase.dev/use-cases/used-part-pricing-from-photos)
HTML
<a href="https://trustcase.dev/use-cases/used-part-pricing-from-photos"><img src="https://trustcase.dev/badge/used-part-pricing-from-photos.svg" alt="Trustcase trust score for Used part pricing from photos" height="20"></a>
  • Amazon listing keyword audit

    Pulls real search suggestions for your product and checks your listing against them. Drafts a new title, bullets and description for you to approve.

  • Store policy contradiction check

    Reads your store's refund, shipping, privacy, terms and contact pages and scores them. It flags missing items and places where a product page or banner promises something the policy does not.

  • Closet inventory from purchase and sale emails

    A weekly run reads your inbox for purchase confirmations and resale item sold notices and drops them in a review tray. Nothing enters or leaves your closet inventory until you confirm it.

  • Supplier stock sync with read back check

    A daily job pulls product and stock data from suppliers, writes it to the store and reads every write back to verify it. Anything that cannot prove it worked fails loudly instead of reporting success.