ARI
Leaderboard

food

Blue Tokai Coffee

bluetokaicoffee.com

shopifycheckout: razorpay_magic
37.8
grade F
ARI score

Agent journey — 15.0/100

Mission: “buy a 250g bag of ground coffee under ₹700

A live AI agent shopping this store — 70% of the ARI score.

Discovery
15.0/15
Product
0.0/20
Cart
0.0/10
4
Checkout
0.0/25
5
Form
0.0/15
6
Payment
0.0/15
BROKEN_FLOWA step in the journey errors or dead-ends.

3 runs · 1 voided as agent error — agent faults are never published as store failures.

Static legibility — 91.0/100

What a machine can read before it even acts — structured data, crawl signposts — 30% of the ARI score.

  • JSON-LD product data
  • llms.txt
  • sitemap.xml
  • OpenGraph tags
  • PDP completeness: 80%
Highest-impact fixes
  1. 1Expose price, availability, and variants as structured data (JSON-LD Product) and as plain DOM text — not just pixels.
  2. 2Make add-to-cart deterministic: stable buttons, a visible confirmation, no silent failures on variant selection.
Reproduced findings
BROKEN_FLOW

A step in the journey errors or dead-ends.

Unadjudicated (done_stage claims repeatedly failed verification): Cart sidebar shows Vienna Roast 250g French Press added at ₹650.

A finding is published only when at least two of three runs reproduce it.

Run log
RunReachedTurnsOutcome
1Product page11BROKEN_FLOW
2Cart22BROKEN_FLOW
3Product page15voided — agent error, not a store finding