What it does
Analyzes completed WooCommerce orders to find products frequently purchased in the same order, then proposes bundle pairings and upsell copy for the product page, rather than guessing at complementary items. Ranks suggestions by how often the pairing occurs and by combined margin where cost data is available.
Install
This skill isn't packaged in an installable registry. It's a plain SKILL.md file you can copy directly. Copy it from the SKILL.md tab and save it at the path for your assistant:
Use it in your AI assistant
Claude Code
.claude/skills/upsell-bundle-suggester/ (or ~/.claude/skills/upsell-bundle-suggester/ for all projects)
GitHub Copilot (VS Code)
.github/skills/upsell-bundle-suggester/
OpenAI Codex
.codex/skills/upsell-bundle-suggester/
Cursor
.cursor/skills/upsell-bundle-suggester/ (or ~/.cursor/skills/upsell-bundle-suggester/ globally)
Compatibility
WordPress with WooCommerce active. Needs read access to order line items, via REST API or database, across a meaningful order history.
The complete SKILL.md for this skill:
---
name: upsell-bundle-suggester
description: Use when a user wants product bundle or upsell suggestions for a WooCommerce store based on real order history rather than guesswork.
---
# Upsell Bundle Suggester
## When to use
A store wants bundle or "frequently bought together" suggestions for one or more products, grounded in actual co-purchase data.
## Procedure
1. Pull a sample of completed orders (at least the last few hundred, or a set time window) and their line items.
2. Count how often each pair of products appears in the same order, and rank pairs by frequency.
3. For the target product, surface its top 2-3 co-purchased items as bundle or upsell candidates.
4. Where cost or margin data exists, note which pairing is highest-margin, not just highest-frequency.
5. Draft short upsell copy for each suggested pairing, one line, focused on the practical reason they go together.
6. If order history is too thin (fewer than a handful of co-occurrences) for a product, say so instead of forcing a pairing.
## Guardrails
- Never invent a "customers who bought this also bought" pairing that isn't backed by actual order data.
- Never claim a discount or bundle price is available unless the user confirms one exists.