API Reference

Complete API reference for @tour-kit/ai: client hooks, server utilities, chat components, and configuration types

Client Exports (`@tour-kit/ai`)

Hooks

`useAiChat()`

Core chat hook. Must be used within AiChatProvider.

Returns UseAiChatReturn:

Property	Type	Description
`messages`	`UIMessage[]`	Chat message history
`sendMessage`	`(input: { text: string }) => void`	Send a user message
`status`	`ChatStatus`	Current chat status (`'ready'`, `'submitted'`, `'streaming'`, `'error'`)
`error`	`Error \| null`	Current error, if any
`stop`	`() => void`	Stop the current generation
`reload`	`() => void`	Regenerate the last response
`setMessages`	`(messages: UIMessage[]) => void`	Replace message history (pass `[]` to clear)
`isOpen`	`boolean`	Whether the chat panel is open
`open`	`() => void`	Open the chat panel
`close`	`() => void`	Close the chat panel
`toggle`	`() => void`	Toggle the chat panel

`useTourAssistant()`

Tour-aware chat hook. Extends useAiChat with tour context.

Returns UseTourAssistantReturn (extends UseAiChatReturn):

Property	Type	Description
`tourContext`	`TourAssistantContext`	Current tour state
`isLoading`	`boolean`	Whether the assistant is loading
`suggestions`	`string[]`	Contextual suggestions based on current step
`askAboutStep`	`() => void`	Ask the AI about the current tour step
`askForHelp`	`(topic?: string) => void`	Ask for help, optionally on a topic
...all `UseAiChatReturn` properties

`useSuggestions()`

AI-generated follow-up suggestions.

Returns UseSuggestionsReturn:

Property	Type	Description
`suggestions`	`string[]`	Current suggestions
`isLoading`	`boolean`	Loading state

Utilities

`SlidingWindowRateLimiter`

Client-side rate limiter using a sliding window algorithm.

`createRateLimiter(config)`

Factory function for creating a rate limiter instance.

`createAnalyticsBridge(config)`

Bridges AI chat events to @tour-kit/analytics.

Server Exports (`@tour-kit/ai/server`)

Route Handling

`createChatRouteHandler(options)`

Creates an API route handler for chat requests.

import { openai } from '@ai-sdk/openai'

const { POST } = createChatRouteHandler({
  model: openai('gpt-4o-mini'),
  context: {
    strategy: 'context-stuffing',
    documents: [{ id: 'doc-1', content: '...' }],
  },
  instructions: {
    productName: 'My App',
    tone: 'friendly',
  },
})