SKU Creation & Naming Standardizer

The problem this kills

Your item master is a mess and everyone knows it. Three people created the same product under three different SKUs. Half the codes follow the 2019 naming scheme, half follow the 2022 one, and nobody can remember which attributes are mandatory. Search returns near-duplicates that should have been one item. Reports are wrong because the same thing is counted twice. Every fix downstream - in purchasing, in receiving, in your reports - traces back to the moment a SKU got created badly.

The root cause is almost never bad people. It is that SKU creation is a free-text field in a spreadsheet or ERP with no rules, no duplicate check, and no review. Anyone can type anything, and they do.

This Implementation Plan gives you a guided front door for new items: pick the attributes, the tool builds the SKU your way, warns you about look-alikes, and won't let it reach the item master until a data steward signs off.

What you'll build

A small, team-only web app that does one job extremely well: create clean, standardized SKUs.

• A guided new-item form where the user picks category, attributes, and size from dropdowns - not free text.
• A SKU generator that assembles the code from your exact naming rules (think category-attribute-size codes) so every SKU comes out consistent.
• A duplicate and near-duplicate checker that compares the new item against your existing item master, using fuzzy matching to catch the look-alikes a plain "exact match" search would miss.
• A steward approval gate: the tool drafts the new SKU, a designated reviewer approves or rejects it, and only approved items are committed.
• A clean CSV export in the exact columns your ERP or system of record expects, so you can load new items without retyping.
• A complete audit trail of who created, reviewed, approved, or rejected every item, and when.

What's inside the Implementation Plan

The plan is a single file you paste into an AI coding agent (Claude Code). It walks the agent through building the whole tool, step by step, each step ending in a ready-to-paste prompt.

The most important part: the plan opens by interviewing you about your business. Before it writes a line of code, the agent asks about your real SKU naming scheme, which attributes are mandatory, how your codes are structured, your typical and peak item-creation volume, who is allowed to approve, and your messy edge cases (kits, variants, private-label, legacy codes). It reflects a short tailored spec back to you, gets your thumbs-up, and only then builds. You get a tool shaped to your conventions - not a generic template you have to bend to fit.

Inside you'll find: the discovery interview, the data model tuned to your attributes, the SKU-generation rules, the fuzzy duplicate-matching logic, the approval workflow, the audit log, email notifications, and the CSV import/export fallback so it works today even with no ERP integration.

The governance it includes (this is the point)

This is not a toy. The plan builds in the controls that make a tool safe to put in front of a team:

• Login so only your team can use it.
• Row-level security so users only ever see their own organization's data.
• A hard human-in-the-loop approval gate - the tool drafts a SKU, a steward reviews it, and nothing is committed to the item master until a person approves.
• Duplicate guards that dedupe on both the generated SKU and an attribute fingerprint, so the same item can't be created twice even through two different paths.
• A complete audit trail of every create, edit, approve, and reject.

Who it's for

• Inventory data stewards who own the cleanliness of the item master.
• Product and category managers creating and classifying new items.
• Purchasing teams who need a SKU to exist before they can buy against it.

If you've ever had to do a SKU-cleanup project, this is how you stop needing one.

You've got this - paste the first prompt and let the interview tailor the rest.