AI Customer Service in Slack

FlowHunt allows you to connect your AI customer service chatbot to Slack. This way, you get to monitor all chatbot conversations and also jump in any time and have the chatbot ping you whenever it needs your help. This is called human in the loop. In other words, it’s when AI lets you know what’s happening and proactively asks you to take over within the chatbot window instead of just giving the users options to contact you themselves.

Once you have the Flow in your Flow Editor, you’ll need to set up a couple more things. We can break this Flow up into three major subflows. First is the customer service chatbot itself. Second is the triggers that let you take over chats via Slack. The last part is just simple quality-of-life additions.

Part 1: The Customer Service Chatbot

This is the large Subflow at the bottom that starts with Chat Input. Here’s what you need to set up:

Set the same Workspace and Channel for all three Slack Send Message components. If you want to, you can choose to send the messages to more than one channel.
Open the Slack Send Message connected to Human Escalation. This component controls the message the bot sends you whenever it’s trying to get your attention. That’s why you should set up a mention to notify your team that human assistance is needed.
Set up the Email Notification Tool to allow the chatbot to send you emails whenever you’re not reachable via Slack.
Set up your Human Assist Tool, allowing the chatbot to transfer users to human live chat as a fallback method. By default, LiveAgent is connected, but you can quickly swap it for any Human Assist Tool.
Set up the AI Agent. Give it a specific role, name, and personality. For example: “You are a helpful AI customer service chatbot representing [company]. Your name is [bot name]. You are [personality traits, type of communication, hobbies, etc.]”.

Part 2: The Slack Triggers

This is the smaller subflow in the top left. Here’s what you need to set up:

Set the Worskpace and Channel in Slack Message Recieved.
Set the same Workspace and Channel in both Slack Send Message.
Feel free to edit the messages in the Prompt Templates. These messegase are arbitrary and only for your convenience. They inform you about the chatbot status, and are only visible in Slack.

Part 3: Welcome Messages and FAQ

This is the smaller subflow in the top right. It’s purpose is to proactively send the welcome message, instructions and FAQ buttons to the user at the start of the chat. Here’s what you need to set up:

Write your welcome message into the Message field of the Message Widget.
If you want to add more messages, simply add another message widget, connecting it to Chat Open Trigger on one side and Chat Output on the other.
Add questions and button options as Button Text in the Button Widget. You can add an optional title to each button.
If you want to add more messages, simply add another message widget, connecting it to Chat Open Trigger on one side and Chat Output on the other.

Components breakdown

Chat Input: Represents the message you send in the chat. It’s the starting point of any flow.
Chat Output: Component representing the chatbot’s answer.
Chat Open Trigger: Activates when a user opens the chat window, enabling actions like welcome messages or predefined behaviors.
Slack Message Received This component represents the start of the Slack conversation with FlowHunt. You can think of it as a trigger component. It also lets you control when, where, and how Flo’s responses are triggered.
Send Slack Message: This component represents Flowhunt’s Slack messages back to you. It lets you control where, how, and to whom Flo sends answers.
Notification Email Tool: Captures contact information of users and sends it it your email.
Human Assist Tool: Allows for transferring users to supported live chat solutions.
URL Retriever: Allows your Flow to retrieve information from individual URLs.
Document Retriever: Allows the bot to search and use information from your website and linked documents.
Chat History: Ensures the bot remembers a set amount of previous messages.
Tool Calling Agent: An autonomous AI with its own reasoning.
GoogleSearch: Searches the web for results relevant to the user query.