ChatbotKit Integration

Overview

ChatbotKit is a comprehensive platform for building, training, and deploying AI chatbots without coding. It offers an intuitive interface for designing conversational flows, integrating knowledge bases, and managing bot deployments.

Note: Custom LLM integration is available only in the Pro, Pro-Plus & Teams version of ChatbotKit.

Key Features

Visual Bot Builder: Create conversation flows with a drag-and-drop interface
Knowledge Base Integration: Train bots with your custom documents and data
Multi-Channel Deployment: Deploy to websites, messaging apps, and custom platforms
Analytics Dashboard: Track performance and user interactions
Conversation Management: Monitor and intervene in live conversations

Use Cases

Customer support automation for websites and products
Internal knowledge base assistants for employees
Lead generation and qualification chatbots
Educational and training assistants
FAQ and information retrieval systems

Setup Instructions

Sign Up for ChatbotKit:
- Visit ChatbotKit’s website
- Create an account (optional: select your subscription tier)
- Log in to your dashboard
- Select Create New Bot to start a new project
- Choose a template or start from scratch by choosing custom from the list of templates.
Configure Bot Settings:
- After choosing the custom template, navigate to the bot configuration page
- Enter a name and description for your bot
- Configure any other required settings, for this example we’ll go with a document dataset for the bot.
- Click on “Advanced Bot Customization options” to access the API settings page.
- Select “OpenAI Compatible API” from the options
- Configure the following settings:
```
Model: Custom
Provider: OpenAI
API Key: your_api_key
API Endpoint: https://api.relax.ai/v1/
```
- Set default parameters for temperature, top_p, etc.
- Click “Save Provider” to store your configuration
For more information, checkout the models guide on ChatbotKit
Create a New Chatbot:
- Return to the Dashboard and click “Create Chatbot”
- Select a template or start from scratch
- Enter a name and description for your chatbot
Select AI Model:
- In the bot settings, navigate to “AI Configuration”
- Choose your custom provider from the dropdown
- Select your preferred model
- Adjust parameters as needed for your use case
Build Conversation Flow:
- Use the visual editor to create your conversation flow
- Add nodes for greetings, questions, answers, and actions
- Connect nodes to create the conversation path
Add Knowledge Base (optional):
- Navigate to the “Knowledge” tab
- Click “Add Knowledge Source”
- Upload documents or provide URLs for the bot to learn from
- Configure retrieval settings for accuracy
Test Your Chatbot:
- Click the “Test” button to open the testing interface
- Try different conversations to verify functionality
- Make adjustments to improve responses
Deploy Your Chatbot:
- Go to the “Deploy” tab
- Choose a deployment method (Widget, API, or Integration)
- Configure appearance and behavior settings
- Copy the provided code or follow integration instructions

Advanced Configuration

ChatbotKit offers several advanced features for complex chatbot scenarios:

Contextual Understanding:
- Configure entity extraction for identifying key information
- Set up context variables to maintain conversation state
- Define conversation memory settings for long-term interactions
Handoff Logic:
- Create conditions for transferring to human agents
- Set up email notifications for specific triggers
- Configure working hours and availability rules
Integrations:
- Connect to CRM systems like Salesforce or HubSpot
- Integrate with help desk software like Zendesk or Intercom
- Set up webhook connections to your backend systems