Output Nodes Reference

Output nodes return results from your workflow, sending responses to users, chat interfaces, or external systems.

---

Text Response Node

Send text messages back to users with advanced templating support.

Configuration

Field	Type	Required	Description
Text	Text	Yes	Response text template
Typing Indicator	Toggle	No	Show typing indicator before response
Delay	Number	No	Delay before sending (seconds)

Template Features

Variable Insertion:

Hello {{contact.first_name}}, your order #{{order.id}} is confirmed.

Array Iteration:

Your items:
{{#each order.items}}
- {{item.name}}: ${{item.price}}
{{/each}}

Conditional Content:

{{#if success}}
Your request was successful!
{{else}}
Sorry, there was an error.
{{/if}}

Accessing Previous Node Output:

{{llmintegration_1.output}}

Output Variables

{{textresponse_1.response_text}}  - Final formatted text
{{textresponse_1.output}}         - For chaining to next node

---

Response Builder Node

Compose rich, multi-component responses with text, tables, charts, SQL, and more.

Configuration

Field	Type	Required	Description
Components	Array	Yes	Array of component configurations
Output Destination	Select	No	Where to send output: chat, sse_output, api_response

Component Types

Type	Icon	Description
text	📝	Rich text with markdown support
sql	🗄	Syntax-highlighted SQL query
table	📊	Interactive data table with sorting/pagination
chart	:chart_increasing:	Plotly chart visualization
recharts	📊	Recharts visualization (bar, line, pie, area)
contacts	👥	Contact cards display
activities	📆	Activity timeline
json	{}	Collapsible JSON tree viewer
alert	⚠️	Highlighted alert/notice message
tableau	📊	Embedded Tableau visualization
image	🖼	Image display with lightbox

Component Examples

Text Component:

{
  "type": "text",
  "props": {
    "content": "## Summary\n\nHere are the results...",
    "maxLength": 500,
    "showHeader": false
  }
}

Table Component:

{
  "type": "table",
  "props": {
    "data": "{{queryNode.results}}",
    "pageSize": 10,
    "showSizeChanger": true
  }
}

Chart Component (Recharts):

{
  "type": "recharts",
  "props": {
    "data": "{{analysisNode.chart_data}}",
    "type": "bar",
    "title": "Sales by Month",
    "xKey": "month",
    "yKey": "revenue"
  }
}

Alert Component:

{
  "type": "alert",
  "props": {
    "message": "Processing complete!",
    "type": "success"
  }
}

Output Variables

{{responsebuilder_1.output}}      - Full response object
{{responsebuilder_1.components}}  - Array of rendered components
{{responsebuilder_1.success}}     - true/false

---

Voice Response Node

Convert text to speech and output audio responses.

Configuration

Field	Type	Required	Description
Text	Text	Yes	Text to convert to speech
Synthesizer Source	Select	Yes	direct, voice_agent, or none
Synthesizer Integration	Select	Conditional	TTS integration (when source is direct)
Output Format	Select	No	mp3, wav, ogg, m4a
Output Destination	Select	No	file, stream, phone, webrtc

Voice Settings

Setting	Type	Description
Voice ID	Text	Voice to use for synthesis
Speed	Number	Speech speed (0.5-2.0, default 1.0)
Pitch	Number	Voice pitch (-20 to 20, default 0)
Volume	Number	Audio volume (0.0-1.0, default 0.8)

Example Configuration

Text: "Hello {{contact.first_name}}, thank you for calling."
Synthesizer Source: direct
Output Format: mp3
Voice Settings:
  Speed: 1.0
  Volume: 0.8

Output Variables

{{voiceresponse_1.audio_url}}     - URL to audio file
{{voiceresponse_1.audio_data}}    - Audio binary data
{{voiceresponse_1.duration}}      - Audio duration
{{voiceresponse_1.success}}       - true/false

---

Workflow JSON Output Node

Validate and output workflow JSON configurations (used by AI Workflow Builder).

Configuration

Field	Type	Required	Description
Workflow JSON	Text	Yes	Workflow JSON from LLM
Auto Fix Handles	Toggle	No	Auto-fix missing sourceHandle/targetHandle
Auto Layout	Toggle	No	Auto-layout nodes using grid
Validation Level	Select	No	strict or lenient
Output Channel	Select	No	workflow_builder or api_response

Use Case

This node is primarily used in AI-powered workflow generation:

[LLM Integration] → [Workflow JSON Output]
                    - Validates generated workflow
                    - Fixes handle issues
                    - Outputs to workflow builder

Output Variables

{{workflowjsonoutput_1.is_valid}}          - Whether workflow is valid
{{workflowjsonoutput_1.workflow_json}}     - Validated workflow JSON
{{workflowjsonoutput_1.total_nodes}}       - Number of nodes
{{workflowjsonoutput_1.total_edges}}       - Number of edges
{{workflowjsonoutput_1.fixes_applied}}     - Number of auto-fixes
{{workflowjsonoutput_1.validation_errors}} - List of errors

---

Quick Reference

Node Selection Guide

Need	Use
Simple text message	Text Response
Rich multi-component output	Response Builder
Voice/audio response	Voice Response
AI-generated workflow output	Workflow JSON Output

Response Patterns

Simple Text:

[Process Data] → [Text Response: "Done! Found {{count}} records."]

Rich Dashboard Response:

[Query Data] → [Response Builder]
               - Text: Summary
               - Table: Results
               - Chart: Visualization

Voice Reply:

[LLM Generate] → [Voice Response: "{{llm.output}}"]

Best Practices

1. Use Text Response for simple messages - Fast and efficient
2. Use Response Builder for rich data - Tables, charts, multiple sections
3. Add typing indicator - Makes responses feel more natural
4. Handle empty data - Check for results before displaying tables/charts
5. Use appropriate delays - Prevent overwhelming users with rapid responses