CurioSetupInboxInsightsFeedbackTrustCatalog gapsAssistantCampaignsBillingProductsKnowledgeTest chatDemo storefront

Sample data — this is not a real store

No Shopify store is connected to this session. Every number on this page — conversations, product counts, ratings — comes from a seeded demo catalog of 65 snowboards, not from a merchant's shop. Open the app from your Shopify admin to see your own store. Treat this store's chat as public: don't enter real personal details.

Trust

How grounded the assistant stays in your catalog and policies.

Last 30 daysFeedback

Sample data — this is not a real store

No Shopify store is connected to this session. Every number on this page — conversations, product counts, ratings — comes from a seeded demo catalog of 65 snowboards, not from a merchant's shop. Open the app from your Shopify admin to see your own store. Treat this store's chat as public: don't enter real personal details.

Curio will not invent

Answers come from your catalog and knowledge. Prices, stock, and policy promises are only what you have indexed — otherwise the assistant refuses or offers a human handoff.

Showing the last 7 days.

Trust score

36

Thumbs-up rate

n/a

Source hit rate

32%

Honest refuse

44%

Score breakdown

Weights: thumbs 40%, source hits on product/policy turns 40%, honest refuse when retrieval is empty 20%. Missing components are skipped and the rest reweighted — we never invent a high score.

Thumbs-up rate

No ratings yet

40% weight

n/a

Source hit rate

44 grounded / 139 product·policy turns

40% weight

32%

Honest refuse (empty RAG)

41 honest / 93 empty-RAG turns

20% weight

44%

Sample: 223 AI turns, 0 ratings, 148 trust signals.

Recent thumbs-down

Open feedback

Fix these in Knowledge or your catalog — they drag the trust score and are the cheapest list of misses you will get.

No thumbs-down yet

When shoppers mark a reply not useful, it shows up here and on Feedback.