How to use it

1. Download the file and put it in a folder called shopify-cross-sell-recommender.
2. Claude: zip the folder and upload it under Settings, Capabilities, Skills. Claude Code, Codex, Gemini CLI or Cursor: drop the folder in your skills directory.
3. Connect your store with the Shopify connector (and GA4 or ad accounts where the skill mentions them).
4. Ask for the job in plain English. The skill loads itself when the request matches.

What's inside

Cross-Sell Recommender

When the user asks for cross-sell recommendations, use the connected Shopify MCP to:

1. Pull orders

Last 180 days of orders. For each: - Order ID - Line items (product IDs + quantities) - Customer ID, order total

2. Build co-purchase matrix

For every pair of products (A, B) that appeared in the same order: - Count co-occurrences - Support = orders(A∩B) / total orders - Confidence = orders(A∩B) / orders(A) - Lift = confidence(A→B) / support(B)

3. Filter to high-affinity pairs

• Confidence ≥ 10% (at least 1 in 10 buyers of A also buys B)
• Lift ≥ 1.5 (B is bought 1.5x more often when A is in cart)
• Minimum 30 co-occurrences (avoid noise)

4. Rank for top products

For each of the top 20 products by revenue: - Top 3 cross-sell candidates sorted by lift × margin of B - Exclude bundles already shown via Shopify Combined Listings

5. Output

PDP-ready table: | Anchor product | Cross-sell 1 | Cross-sell 2 | Cross-sell 3 | Expected AOV lift |

Then the exact Shopify product metafield format for cross-sell apps (e.g., Rebuy, Boost AI). Always test on 1-2 PDPs before rolling out site-wide.

Also on GitHub.