What are Variable Inputs?
Variable Inputs allow your AI agents to collect specific information from users before starting a conversation. This creates a more personalized and effective experience by ensuring your agent has all the context it needs from the beginning.
Benefits for Your Team
Personalized Conversations: Collect client IDs, project names, or preferences
Context-Aware Responses: Agents understand the specific scenario from the start
Streamlined Workflows: Eliminate repetitive questions at the beginning of each chat
Consistent Data Collection: Ensure all required information is gathered in a structured format
How It Works
For Agent Creators
Define Variables: Use the format {{variable_name}} in your agent's system prompt
Configure Input Types: Customize how each variable is collected (text, dropdown, number, date)
Set Validation Rules: Ensure users provide valid data (required fields, number ranges, patterns)
For End Users
Before chatting with an agent, users see a form with the required input fields
After filling in the information, the conversation begins with all context pre-loaded
The agent immediately uses this information in its responses
Setting Up Variable Inputs
Step 1: Define Variables in Your System Prompt
Include variables in your agent's system prompt using double curly braces:
text
You are a support agent helping with {{product_name}} issues.
The customer's account type is {{account_tier}} and they're using version {{version_number}}.Step 2: Configure Variables in the Agent Settings
Navigate to your agent settings
Go to the "Variable Inputs" tab
Any variables from your system prompt will be automatically detected
Click on each variable to configure:
Display Name: How it appears to users
Input Type: Text, dropdown, number, or date
Required: Whether the field is mandatory
Default Value: Pre-populated information
Description: Help text for users
Validation: Rules to ensure valid input
Step 3: Configure Input Types
Text Inputs
Simple text fields for names, IDs, or free-form information
Optional pattern validation for formats like email or phone numbers
Dropdown Inputs
Static Options: Define a fixed list of choices
API Source: Dynamically load options from an API endpoint
Number Inputs
Set minimum and maximum values for numerical data
Date Inputs
Calendar picker for selecting dates
User Experience
When someone starts a chat with your agent:
They'll see your agent's profile and description
Below that, a form will appear with all required inputs
After submitting valid information, the conversation begins
The agent uses these inputs throughout the conversation
Advanced Configuration
Dropdown Options
For dropdown inputs, you can:
Create a static list of options with labels and values
Connect to an API endpoint that returns dropdown options
Map API response fields to option labels and values
Validation Rules
Ensure data quality with validation:
Required Fields: Mark fields that cannot be skipped
Pattern Validation: For text that should follow specific formats
Number Ranges: Set minimum and maximum values
Custom Error Messages: Show helpful guidance when validation fails
Best Practices
Writing Good Variable Names
Use clear, descriptive names: client_id instead of cid
Use underscores for multi-word variables: project_name instead of projectName
Stay consistent with naming conventions across agents
Creating User-Friendly Forms
Keep the number of required fields minimal
Provide clear descriptions for what information is needed and why
Use appropriate input types for each variable
Group related variables together
Instructions/System prompt
Reference variables naturally in your prompt text
Use variables to create conditional instructions for the agent
Ensure variable substitution makes grammatical sense in your prompt
Example Use Cases
Customer Support Agent
Collect account ID, issue category, and urgency level
Agent immediately knows customer context and prioritization
Sales Assistant
Gather industry, company size, and current tools
Agent can provide tailored product recommendations
Project Management Bot
Collect project name, team size, and timeline
Agent can suggest appropriate methodologies and tools
Analytics Assistant
Specify data source, time period, and metrics
Agent can immediately generate relevant reports
Troubleshooting
Variable not appearing in form?
Verify the variable is correctly formatted in your system prompt as {{variable_name}}
Check that you didn't accidentally delete it in the Variables tab
Dropdown options not loading?
Ensure your API endpoint is accessible and returns the correct format
Verify your label and value mappings match your API response structure