FlowRunner
PricingContact
Theme
Start Free

LiveChat

Communication

Manage LiveChat conversations from your flows. Agents read active chats, post messages and internal notes, start proactive outbound chats, close resolved conversations, and create or list agents on the account.

7 actions Basic auth available
LiveChat website ↗ Platform Documentation ↗ Capability data verified 2026-07-28
A new support conversation appears in the queue during peak hours
Get Chat pulls the full thread, participants, and chat properties
The agent drafts a reply from the account history and posts it into the chat with Send Message set to agents-only visibility
The draft is scored against policy, and anything touching refunds, commitments, or account changes is held for review
The on-duty supervisor reads the internal note alongside the customer's messages and approves or edits the reply
Send Message posts the approved text into the chat where the customer sees it
Deactivate Chat closes the conversation once the customer confirms the issue is resolved
A transcript summary lands on the CRM record and in the daily support log

What This Integration Enables

Live chat is the highest-stakes writing surface in the support stack: a real customer, in real time, with the clock visible to both sides. The pressure to answer fast is exactly what produces the wrong answer. FlowRunner splits the work along that line. Agents do everything speed rewards, reading the thread with Get Chat, pulling account context, drafting the reply, logging the outcome, while the message that actually reaches the customer can pass through a person whenever the stakes say it should. LiveChat's visibility control makes the split native: the same Send Message action posts either an internal note or a customer-visible reply. - Draft first responses as agents-only notes with full conversation and account context - Post approved replies into active chats and close resolved conversations with Deactivate Chat - Start proactive outbound chats from billing, onboarding, and usage signals - Sync transcripts and outcomes into the CRM, spreadsheets, and notification channels - Provision new support agents on the account as the team grows

Without FlowRunner

First response depends on staffing The customer waits as long as it takes for someone to be free, and peak hours are exactly when nobody is
Context dies in the archive Resolved chats hold decisions and promises the CRM never hears about
Proactive outreach never happens Nobody starts a chat with the account whose payment just failed, because nobody is watching for it

With FlowRunner

Drafts ready before a person picks up Replies arrive as agents-only notes with the full history already read and summarized
Every chat lands where the business looks Transcripts are summarized into the CRM and the reporting sheet the moment a chat closes
Outbound chats from real signals Start Chat opens conversations from billing events and usage cliffs, seeded with the right first message

Use Case Scenarios

The first response that is fast and right

A customer opens a chat about a billing discrepancy. The agent finds the conversation with List Chats, reads it with Get Chat, pulls the invoice history, and posts a proposed reply as an agents-only note: the discrepancy is real, a credit is warranted. Because the draft commits money, the supervisor reviews it in seconds rather than researching it in minutes, approves, and the customer-visible Send Message goes out. The customer experienced one fast, correct answer. The team experienced a review, not a scramble.

The archive that feeds the CRM

When a conversation is resolved and closed with Deactivate Chat, the agent pulls the final transcript with Get Chat, summarizes the issue, resolution, and any commitments made, and writes it to the contact record in [HubSpot](/integrations/hubspot). New customers discovered mid-chat become CRM contacts on the spot, and [Slack](/integrations/slack) alerts the sales team when a support conversation surfaces expansion interest. Reporting rows accumulate in [Google Sheets](/integrations/google-sheets) without anyone exporting an archive.

The chat that opens before the ticket exists

A payment fails twice for a mid-tier account. Instead of waiting for the customer to notice and arrive angry, the workflow calls Start Chat seeded with a specific first message about the card on file, and posts the supporter context as an internal note so whoever picks up the thread knows exactly why it exists. The same pattern reaches out when onboarding stalls at the same step for three days. Outreach happens at the moment of the signal, not at the next campaign.

Human-in-Loop Highlight

Send Message with customer visibility is a write that cannot be unread. In a live chat, the customer is watching the typing indicator, and whatever posts is instantly part of the relationship: the refund promised, the discount implied, the policy misstated. So FlowRunner flows exploit the action's own visibility control as a gate. The draft posts as an agents-only note first: "Customer disputes the March invoice. Records support them. Proposed reply credits $140 and apologizes for the double charge. Approve?" The supervisor approves or edits, and only then does the customer-visible message post. The [human-in-the-loop](/concepts/human-in-the-loop) moment lives inside the same conversation, on the same action, one visibility flag apart.

Agent processes routinely
Detects exception requiring judgment
Clear match Continues automatically
Ambiguous Routes to human via preferred channel
Human decides
Agent resumes with decision

Agent Capabilities

7 actions

Conversations

2
  • List Chats Lists archived and accessible chats with the last thread of each, narrowed by filters such as agent, group, or date range, and paged by returned page IDs. The queue view for routing and reporting flows.
  • Get Chat Retrieves a chat's threads, events, participants, and properties by ID, defaulting to the latest thread. The context step before any reply and the transcript source after resolution.

Messaging

3
  • Send Message Posts a text message into an active chat as an agent, with visibility controlling whether everyone or only agents can see it. One action, two jobs: internal drafts and customer-facing replies.
  • Start Chat Starts a new chat as an agent, optionally seeded with an initial message and created as active. The proactive outreach primitive.
  • Deactivate Chat Closes an active chat by ID, ending the thread and moving the conversation to the archive. The clean end-state for resolved conversations.

Team

2
  • List Agents Lists agents on the account with ID, name, role, and status, filterable by group. Used to route conversations and audit coverage.
  • Create Agent Creates a new agent with email, display name, and role, pending their invitation acceptance. The onboarding step when the support team grows.

Frequently Asked Questions

What can FlowRunner do with LiveChat?

FlowRunner agents can run List Chats, Get Chat, and Send Message in LiveChat, plus 4 more actions.

Does connecting LiveChat to FlowRunner require OAuth?

No. LiveChat connects to FlowRunner with basic authentication (a username and password), no OAuth flow required.

Can LiveChat trigger a FlowRunner workflow automatically?

LiveChat doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.

Start building with LiveChat

$100 in credits. No card required. Connect in minutes.