Slack AI Integration for Customer Support & Lead Alerts

It connects your AIChatVault chatbot with your Slack workspace in three ways: people can ask the bot questions by mentioning it or sending it a direct message, your team receives alerts for qualified leads and human-handoff requests, and your team can take over, reply and resolve those conversations from Slack.

Mention the bot in a channel it has been added to, or send it a direct message. In a channel it replies in a thread; in a direct message it replies inline. It does not read or answer every message in a channel — only mentions and direct messages.

The same chatbot you already built. It answers from the knowledge sources you added — websites, documents and Q&A — and, when a store is connected, from live product information. If the knowledge base does not contain the answer, it says so instead of guessing.

Yes. Each AI answer uses credits in the same way a website chat message does, counted against your plan. Messages answered without the AI — such as a handover to your team — do not use credits.

Two: a qualified lead is captured, and a human handoff is requested. Both are triggered by conversations on any connected channel — website chat, WhatsApp, Instagram, Messenger, email or Slack itself.

Each lead gets a score from 0 to 100 based on the details captured and the conversation: an email address, a phone number, a name, a company, a detailed message, buying-intent wording, a handoff request and an active conversation all add points. A Slack rule sends an alert only when the score reaches the minimum you set. The default minimum is 50.

Yes. Create separate rules — up to 20 per chatbot — each with its own event, Slack channel, minimum lead score and the details to include: customer information, a summary, a link to the conversation, the lead score, the page visited and the source channel.

Take ownership of the conversation, open it in AIChatVault, or resolve it, and the alert updates in place to show who did what. Conversations that started in Slack can also be answered straight from the alert. Conversations from other channels — website chat, WhatsApp, Instagram, Messenger and email — are answered from the shared AIChatVault inbox, and the Slack alert follows along.

In Slack, a clear request such as “connect me with a human” hands the conversation to your team straight away: the customer is told a teammate has been notified, the AI stops answering that conversation, a lead is recorded and your handoff alert is sent. On other channels, your chatbot’s handoff behaviour decides when the alert is raised; once a teammate takes over, the AI stays out of that conversation.

Yes. When your team takes over, replies or resolves from the inbox, the Slack alert for that conversation is updated too, so both places stay in step. Replies written in the inbox to a Slack conversation are posted back into that Slack thread.

Open your chatbot, go to Integrations → Slack and choose Connect Slack. Approve the connection in your workspace, pick a default channel, add your alert rules and send a test alert. Invite the bot to any private channel you want it to use.

Permission to read mentions and direct messages, post messages, see channel and workspace names so you can choose where alerts go, and read basic user names. Reading user email addresses is optional and only needed to match Slack people to teammates when they take over.

Yes. Pause the whole integration or switch off AI replies while keeping alerts, at any time. Disconnecting removes the stored access token.

Every request from Slack is verified with Slack’s signature and rejected if it is invalid or too old. Your workspace access token is stored encrypted, and each chatbot connects its own workspace.