This is an example implementation of assistant-ui with Inconvo as an external runtime provider.
First, create a .env and file to add your Inconvo API key:
touch .envThen update .env with your credentials:
INCONVO_API_KEY="your_inconvo_api_key_here"
INCONVO_API_BASE_URL="https://app.inconvo.ai/api/v1"
Then install dependencies:
npm iThen, run the development server:
npm run dev
# or
yarn dev
# or
pnpm dev
# or
bun devOpen http://localhost:3232 with your browser to see the result.
app/InconvoRuntimeProvider.tsx- Inconvo runtime integration for assistant-uiapp/assistant.tsx- Main assistant UI componentapp/api/inconvo/- API routes for Inconvo integrationcomponents/inconvo/- Custom message components for charts and tables
The custom runtime provider that integrates Inconvo with assistant-ui, handling:
- Message streaming
- Conversation persistence
- Custom message types (charts, tables)
This example supports rendering:
- Charts: Visualize data using Chart.js
- Tables: Display structured data in table format
You can start editing the assistant by modifying:
app/assistant.tsx- Main UI componentapp/InconvoRuntimeProvider.tsx- Runtime behaviorcomponents/inconvo/- Custom message renderers