support.js
// npm install @bookbag/sdk
import { Bookbag } from '@bookbag/sdk'
const bookbag = new Bookbag({ apiKey: process.env.BOOKBAG_API_KEY })
// Send a customer message to your AI agent and get a reply
const { reply, resolved, handoff } = await bookbag.chat({
agentId: 'support-agent',
conversationId: 'conv_123',
message: 'Where is my order #1042?',
customer: { email: 'sam@example.com' },
})
console.log(reply) // grounded answer with live order status
console.log(resolved) // true if the agent fully resolved it
console.log(handoff) // set when the agent escalates to a humanWhat you can build
Chat API
Send customer messages and get grounded, action-taking replies for any surface — your app, kiosk, or backend.
Custom actions
Let the agent call your own endpoints (loyalty, warranty, OMS) with auth and guardrails you define.
Webhooks
React to handoffs, resolutions, CSAT, and tags in real time to sync with your stack.
Knowledge sync
Programmatically add and refresh knowledge sources so the agent always answers from current content.
Widget JS API
Open/close the widget, prefill customer identity, and pass page context for smarter answers.
Scoped keys
Issue scoped API keys per environment, with test keys and a sandbox agent for safe development.
Frequently Asked Questions
Start building with Bookbag
Join the ecommerce teams resolving more tickets, answering 24/7, and turning support into a revenue channel with Bookbag.