Cartonization & Box-Size Helper

The problem this kills

Your packers grab whatever box is on the shelf. A small order goes out in a giant carton stuffed with air, and your carrier charges you for the dimensional weight - the size of the box, not what's actually in it. You pay to ship air, every single day, and nobody can see how much it's costing until the invoice lands.

Meanwhile the "right box" lives in someone's head. New packers guess. Oversized cartons burn through void fill. And when finance asks "could we have used a smaller box on these 4,000 shipments?" there's no log to answer with.

What you'll build

A simple internal web tool your packing team logs into. They load an order's items, and the tool recommends the smallest box (or boxes) that the items actually fit in, estimates how much void fill you'll need, and shows the dimensional-weight cost estimate before the package is sealed. The packer confirms the suggested box or picks a different one, and every decision is written to a clean pack/box log you can export as CSV for rate shopping and carrier negotiations.

It uses a fast, approximate 3D fit (good enough for real packing decisions), and it loudly flags any item that won't fit in any box you own, so nothing silently slips through.

What's inside the Implementation Plan

• A copy-paste runbook you hand to an AI coding agent (Claude Code) - no coding required from you.
• It opens by interviewing you about your business - your current packing process, your box catalog and inner dimensions, how your item dimensions are named and stored, your carrier's dim-weight divisor, your typical and peak order volumes, and your messy exceptions. It tailors the tool to how you actually pack, instead of dropping a generic template on you.
• Step-by-step build prompts you paste in one at a time - each ends with a ready-to-copy prompt.
• The fit-and-recommend logic, the void-fill and dim-weight estimates, the packer confirm/override screen, and the CSV export - all built and explained in plain language.
• A "No API yet?" fallback so you can build the whole thing today from spreadsheets, even with zero integration to your shipping system.

The governance it includes (this is the point)

This isn't a toy. The plan builds in the controls that make a tool safe to use in a real warehouse:

• Login so only your team can use it.
• Row-level security so each organization only ever sees its own boxes, items, and logs.
• A complete audit trail - who recommended, confirmed, or overrode which box, and when.
• A human-in-the-loop approval gate: a shipping lead approves the box catalog and dim-weight rules before they go live, and the packer confirms the box before pack-complete. The tool drafts; a person commits.
• Duplicate guards so the same order + box recommendation can't be logged twice.

Who it's for

Packers who want the right box without guessing. Shipping cost analysts hunting dim-weight waste. Fulfillment leads who need a defensible log for carrier rate negotiations. If you run a pack bench, this is for you.

You've got this - paste the first prompt and let the agent interview you.