Skip to main content

For Merchants

Paste a Shopify product link and see how to improve the page.

TL;DR: Start with the Recommendation Audit. Paste a Shopify product link and Colter tells you how to improve the page. Add a buyer question only if you want to compare a real AI answer later.

Start The Audit

Run Recommendation Audit

Colter reads the public page and tells you what to improve first. The audit is read-only and does not change the store. A later answer comparison uses an optional saved buyer question with the same named AI tool under materially unchanged conditions.

Supporting Technical Check

No install:

npx @getcolter/cli check https://your-store.com

Global install:

npm install -g @getcolter/cli
colter check https://your-store.com

What The Legacy Store Score Means

The technical Check score describes Colter's public protocol-evidence rules. It does not show whether an AI tool found or recommended a product.

ScoreMeaning
75+Strong readiness
60-74Basic readiness
40-59Needs work
0-39Not ready

Key dimensions:

  • discovery
  • transaction
  • security
  • ecosystem
  • content quality

Fix The Biggest Gaps

Preview fixes:

colter fix https://your-store.com --dry-run

Write files:

colter fix https://your-store.com --apply --output-dir ./colter-fixes

Typical improvements include JSON-LD, llms.txt, protocol manifests, and crawler posture updates.

See What Agents Experience

colter test https://your-store.com --models claude --json

Hosted Test is available on Pro and Agency.

Monitor Live Traffic

colter lens https://your-store.com
colter lens status --url https://your-store.com --period 7d

The technical monitoring view can show recorded:

  • request or browser classifications
  • page paths and checkout reach
  • protocol observations
  • order events matched to available activity records

A recorded or matched event does not show that a visit caused an order or revenue.

Shopify Note

Shopify may provide parts of the protocol surface. Use Colter's technical tools to record only the public evidence or runtime behavior they actually observe.

Next Steps