AI Request
Send a prompt to the AI and get a response. This is the core action for AI-powered workflows.
Configuration
| Field | Description |
|---|---|
| Prompt | Instructions for the AI |
| Return Result to Chat | Toggle ON to send response directly as message |
| Use Structured Output | Toggle ON to extract specific data |
Basic Usage
- Add an AI Request node
- Write your prompt: "Answer this customer question based on the message"
- Enable Return Result to Chat
The AI response is sent directly to the user.
Structured Output
Extract specific information as variables for use in next nodes.
How to Configure
- Enable Use Structured Output
- Click Add Field
- For each field:
- Field Name: Variable name (e.g.,
title) - Type: Text, Number, etc.
- Description: Help AI understand what to extract
- Field Name: Variable name (e.g.,
Example: Support Ticket Extraction
| Field | Type | Description |
|---|---|---|
title | Text | Support ticket title |
priority | Text | high, medium, or low |
category | Text | billing, technical, general |
After the node runs, use these in next nodes:
aiResult.titleaiResult.priorityaiResult.category
Use Cases
- Generate AI responses to user messages
- Extract information from conversations
- Classify message intent or sentiment
- Summarize long conversations