Skip to content

Rich content trust policy

Structured cards and carousels in Chatevo follow a strict trust policy: interactive offers, prices, and media appear only when they come from a successful tool execution mapped by result_transform. The assistant cannot fabricate commerce UI from documents or imagination.

AllowedNot allowed
Cards from HTTP 2xx tool JSON + valid transformCards from failed or timed-out tools
Prices/images from your API fieldsLLM-invented product lists
Actions you define in transform configFake “Buy now” links to unrelated URLs

Visitors treat carousel cards like real inventory. Showing unverified offers would mislead customers and create compliance risk. Chatevo builds RICH_CONTENT on the server after tool success — the model never authors card payloads directly.

Tool outcomeChat shows
Success + transform matchCards + assistant message
Success + transform mismatchText only (check transform reference)
4xx/5xx or network errorText apology; no cards
Tool not calledNo cards for that data
  • Always attach tools to real APIs for catalog/offer UX.
  • Use Test a tool with production-like JSON before go-live.
  • Do not rely on the master prompt to “show products” without a tool.

See also Rich content trust (security) for org-wide policy context.