JSON Block Node

The JSON Block node stores and outputs static JSON schemas or data structures. This utility node is essential for defining data schemas, providing template structures for JSON Builder nodes, storing configuration objects, and maintaining reusable JSON structures throughout your workflows. It acts as a JSON data source or schema repository within your flow.

---

Basic Usage

Use the JSON Block node to store JSON schemas, configuration objects, or structured data that can be used by other nodes in your workflow.

---

Inputs

The JSON Block node does not accept inputs from other nodes. All JSON content is defined within the node's configuration.

---

Outputs

• Output: The JSON content stored in the node, which can be used as a schema definition, data structure template, or configuration object by downstream nodes.

---

Configuration

JSON Content Area

Large Text Input Field: Enter your JSON content directly in the provided text area.

• Supports full JSON syntax including objects, arrays, nested structures
• Can contain schemas, data templates, or actual data
• Supports multi-line formatting for readability
• Must be valid JSON format (though validation may be lenient)

Example JSON Content:

{
  "type": "object",
  "properties": {
    "employees": [
      {
        "name": "string",
        "id": "string",
        "email": "string",
        "age": "number",
        "status": "string"
      }
    ]
  }
}

---

Example Workflows

Employee Data Schema for JSON Builder

Scenario: Define a reusable JSON schema for extracting employee information from text reports using the JSON Builder node.

Steps to Create the Flow:

1. Add a Flow Call Receiver node to receive input from a parent flow:

   • Configure slot-1 to receive report text
   • This acts as the sub-flow entry point

2. Add a Text node with the source data:

   • Contains the employee report text to be processed
   • Example content:

   Company Employee Report 2025
   Employee #1: John Smith (ID: EMP001, john@company.com, age 32, active) works as Senior 
   Developer in Engineering department.

3. Add and connect a JSON Block node:

   i. Define the JSON schema:

   • Click in the JSON content area
   • Enter your schema definition:

   {
     "type": "object",
     "properties": {
       "employees": [
         {
           "name": "string",
           "id": "string",
           "email": "string",
           "age": "number",
           "status": "string",
           "position": "string",
           "department": "string"
         }
       ]
     }
   }

   ii. This schema defines:

   • The overall structure (object with employees array)
   • Field names (name, id, email, age, status, position, department)
   • Data types for each field (string, number)

4. Add a Text node with extraction instructions:

   • Provides instructions for the JSON Builder:

   Extract information for all employees mentioned in the report. Create an array of employee 
   objects and provide summary statistics.

5. Add a JSON Builder node:

   • Connect Text node (report data) → Input
   • Connect JSON Block output → JSON Schema
   • Connect Text node (instructions) → Overwrite System Prompt
   • The JSON Builder uses the schema to structure the extracted data

6. Add a Flow Call Return node:

   • Connect JSON Builder's JSON Output → Returning Slot Json
   • Returns structured data back to parent flow

Preview:

JSON Block defines the schema:

{
  "type": "object",
  "properties": {
    "employees": [
      {
        "name": "string",
        "id": "string",
        "email": "string",
        "age": "number",
        "status": "string"
      }
    ]
  }
}

JSON Builder uses this schema to produce:

{
  "employees": [
    {
      "name": "John Smith",
      "id": "EMP001",
      "email": "john@company.com",
      "age": 32,
      "status": "active",
      "position": "Senior Developer",
      "department": "Engineering"
    }
  ]
}

Result: The JSON Block provides a reusable, maintainable schema that ensures consistent data structure across multiple uses of the JSON Builder, making it easy to update the structure in one place.

---