Creating Integrations
Step-by-step guide to creating and configuring your first integration
Creating an Integration
Navigate to Integrations
From your organization dashboard, click on Integrations in the left sidebar.
Create New Integration
Click the Create Integration button in the upper right corner.
Select Type
Choose the type of integration you want to create:
- Web Chat: Embeddable chat widget for your website
- API Endpoint: Direct API access for custom integrations
- Discord Bot: Integration with Discord servers
- Slack Bot: Integration with Slack organizations
- Microsoft Teams: Integration with Microsoft Teams
Basic Configuration
Enter the following information:
- Name: A descriptive name for your integration
- Description: (Optional) A brief description of its purpose
- Icon/Avatar: Upload a custom icon or choose from available options
Connect Knowledge Bases
Select one or more knowledge bases to power your integration:
- Click Add Knowledge Base
- Select from your available knowledge bases
- Set priority levels if using multiple knowledge bases
Configure Settings
Adjust various settings according to your needs:
- AI Model: Select the underlying AI model (e.g., GPT-4, Claude, etc.)
- Language: Set the primary language for responses
- Message History: Configure how many previous messages to maintain as context
- Maximum Response Length: Set character limits for responses
Save
Click Create to finish setting up your integration.
Configuration Options
AI Model Selection
Choose the right AI model for your use case:
Model | Best For | Response Speed | Cost |
---|---|---|---|
GPT-4 | Complex queries, high accuracy | Slower | Higher |
GPT-3.5 Turbo | General purpose, balanced | Fast | Lower |
Claude | Long-form content, analysis | Medium | Medium |
Knowledge Base Priority
When connecting multiple knowledge bases:
- High Priority: Primary sources that should be searched first
- Medium Priority: Secondary sources for broader context
- Low Priority: Fallback sources for edge cases
Response Configuration
Fine-tune how your integration responds:
- Response Length: 50-2000 characters
- Temperature: Controls creativity (0.1 = focused, 0.9 = creative)
- Context Window: Number of previous messages to consider (1-10)
Next Steps
Once your integration is created:
Deploy Your Integration
Choose how to deploy your integration to users
Customize Behavior
Fine-tune prompts and response templates
Monitor Performance
Track conversations and optimize performance
Advanced Features
Explore trigger buttons and advanced options
Best Practices for Setup
- Start Simple: Begin with one knowledge base and expand gradually
- Test Early: Use the built-in chat interface to test before deploying
- Clear Naming: Use descriptive names that indicate the integration’s purpose
- Appropriate Models: Choose AI models based on your accuracy vs. speed needs
- Monitor Usage: Keep track of API usage and costs from day one