For chat apps and AI assistants
Give your chatbot answers people can see
Your LLM is great at conversation. For the weather, a recipe or the calories in lunch, WebKnox hands your app a sourced card, and your model keeps everything else.
Only first thing. The overnight rain in Dresden eases off by about 7 am; the rest of Thursday stays mostly dry, between 8 and 12°.
Same question, two answers
A medium banana (118 g) has about 105 kcal, mostly from carbohydrates: 27 g, including 14 g of sugars. It also gives you 3.1 g of fiber and 422 mg of potassium.
Where it fits in your backend
One extra call before your model runs. Nothing to migrate, no session to manage.
- User asksin your chat UI
- Your serversends the messages to WebKnox
- matched: truerender the card, keep its text in history
- matched: falseyour LLM replies as it does today
// server.js: your chat endpoint const res = await fetch('https://api.webknox.com/v2/answers', { method: 'POST', headers: { Authorization: `Bearer ${process.env.WEBKNOX_API_KEY}`, 'Content-Type': 'application/json', }, body: JSON.stringify({ messages }), }); const answer = await res.json(); if (answer.matched) return { type: 'card', answer }; // <webknox-answer> renders it return { type: 'text', text: await myLlm.reply(messages) };
Good to know for chat apps
Follow-ups are chips
“And tomorrow?” arrives as a webknox-follow-up event. Send it as the next user message.
Keep the text in history
Store the answer's text as the assistant turn, so your model and WebKnox both understand “a large one” later.
Match your app's look
Pick a style, set your accent colour, or override any --wk-* property from your CSS.