Trinesis × Amphenol
Secure preview
Please sign in to view this content.
T Trinesis × Amphenol
Back to presentation
Chatbot POC · Integration research

Embedding the AI Assistant into LiveChat

Feasibility & architecture for making the Technical Support Assistant the first-line agent inside Amphenol's existing website live chat — with seamless escalation to a human engineer.

Verdict · Feasible
A first-class, documented use case — not a workaround.

Amphenol uses LiveChat.com (developer platform now branded “Text”, platform.text.com). Our self-hosted FastAPI multi-agent backend plugs in as a native Bot Agent — it answers from our own LLM (not ChatBot.com's no-code builder) and, on request or low confidence, hands the live conversation to a human engineer via transfer_chat with full context retained.

25 / 25
primary-doc claims verified · 0 refuted
Moderate
build effort — no new vendor
Zero
website changes — native widget stays
Self-hosted
our LLM, our backend, our data

Bot-first, human-on-demand

Exactly the model Amphenol asked for: the AI answers instantly, a human steps in the moment it's needed.

Bot answers first

Every new chat fires a webhook; our backend runs the multi-agent LLM (grounded in FitMate data + verified sources) and replies as the bot agent — instantly, 24/7. Same engine as the standalone chatbot POC, just a different transport.

Escalate seamlessly

On an explicit “talk to a human” or low model confidence, the bot calls transfer_chat to reassign the live conversation to a human engineer or group — no restart, full context retained.

Build for: self-messages

incoming_event fires for any event in a chat (including other agents/bots) — the backend must filter by author / visibility so the bot never replies to itself.

Build for: no agent free

A handoff only succeeds if a human is available and accepting chats. We implement a fallback path (e.g. “our team is offline — leave your email”) when no agent is free.

Recommended architecture

The customer keeps LiveChat's native widget — no website rebuild. Only the agent side is our custom bot.

Customer ──▶ LiveChat Chat Widget (unchanged, stays on amphenol-cs.com) │ ▼ LiveChat / Text Platformwebhooks (HTTP POST): incoming_chat, incoming_eventTrinesis FastAPI backend ── runs the multi-agent LLM ── replies via Agent Chat API · send_event (attributed to bot via X-Author-Id) │ └── on “talk to a human” / low confidence ──▶ transfer_chat ──▶ human agent or group
The load-bearing design rule: a bot cannot open a WebSocket (RTM) connection to the Agent Chat API. A custom bot must use the Web API (HTTP request/response) + webhooks. So the backend is a webhook receiver + Web-API caller — not a persistent socket. (The popular narolski/livechat-bot GitHub sample uses RTM and is now disallowed — do not copy it.)

The concrete pieces — all doc-verified

NeedMechanism
Register the botConfiguration API → create_bot_agent (name, status, group priority, webhooks)
Receive customer messagesWebhooks: incoming_chat (new chat), incoming_event (each message) — HTTP POST, return 200, retries on failure
Reply as the botAgent Chat API send_event (message event) with X-Author-Id = bot identity
Escalate to a humantransfer_chat → target type agent or group; confirmed via chat_transferred push
AuthenticationOAuth 2.1. Authorization Code Grant for a private server-side app; Personal Access Token for simpler service access; Client Credentials for a headless bot identity
Customer UILiveChat's native widget stays (single script tag). Only the agent-side bot is custom — no site rebuild

It generalizes (if the vendor ever changes)

The same bot-first → human-handoff pattern is supported by the other majors — so the approach isn't locked to one vendor.

VendorCustom bot + handoffMechanism
LiveChat.com (in use)YesBot Agent · Agent Chat API · transfer_chat
IntercomYesFin AI + Canvas Kit for custom bots
ZendeskYesSunshine Conversations bot-to-agent handoff
SalesforceYesAgentforce / Omni-Channel
FreshchatYesBot-conversation assignment to agents

Primary sources

LiveChat / Text official developer docs — Messaging API overview · Agent Chat API reference · RTM pushes (incoming_event / chat_transferred) · LiveChat apps & webhooks guide · Authorization (OAuth 2.1) · EU data center · Data Processing Addendum.
Verified 01 Jul 2026 — 25 claims cross-checked against primary documentation, 0 refuted.
© Trinesis — LiveChat integration feasibility for Amphenol ← Presentation · Chatbot POC →