Convert

The Convert Node allows you to transform structured data into various file formats. The node converts raw data (e.g. JSON string, etc.) into base64-encoded file data suitable for file storage, downloads, or API responses.

Operation	Description
Convert JSON String to JSON (Base64)	Converts structured data into a pretty-printed JSON file
Convert JSON String to CSV (Base64)	Converts an array of objects into a CSV file with optional header row
Convert Text/JSON String to TXT (Base64)	Converts text or JSON into a plain text file
Convert Text/JSON String to DOCX (Base64)	Converts text or structured data into a Word document with tables

How Does It Work?

Setup Steps:

Build your workflow using Graph Edit Mode (minimum setup: input node → convert node → output node, but you can add more nodes as needed)
Click on the convert node settings to open its configuration
Configure the following data:
- Select the operation (JSON, CSV, TXT, or DOCX)
- Enter raw data or use template variables (e.g., {{{ input_data }}}). Refer to each operation’s Input Format section for accepted data types
- For CSV conversion: Choose whether to include a header row

Output:

The Convert Node returns the converted file data in the following format:

{
  "data": "<<base64-encoded-file-data>>"
}

The data field contains the base64-encoded file data based on the configured conversion operation. This encoded data can be:

Uploaded to cloud file transfer
Passed to other systems that accept base64 file data

Note: The convert node will always return the variable name as data.

Operations

Convert JSON String to JSON (Base64)

Input Format

Input must be a valid JSON string or object

Examples

Example 1 — JSON array:

// Input
[
  {
    "name": "John",
    "age": "25"
  },
  {
    "name": "Jane",
    "age": "30"
  }
]

// Output
{
  "data": "WwogIHsKICAgICJuYW1lIjogIkpvaG4iLAogICAgImFnZSI6ICIyNSIKICB9LAogIHsKICAgICJuYW1lIjogIkphbmUiLAogICAgImFnZSI6ICIzMCIKICB9Cl0="
}

Convert JSON String to CSV (Base64)

Input Format

Data must be an array of objects
All objects must have the same property keys
Values must be simple types (string, number, boolean, null). Nested objects or arrays are not supported
Headers are sorted alphabetically in the output

Valid input:

[
  {"name": "John", "age": 25, "city": "New York"},
  {"name": "Jane", "age": 30, "city": "Boston"}
]

Invalid input:

[
  {"name": "John", "details": {"age": 25}}  // ❌ Nested object not allowed
]

Additional setting: Include Header Row — Checked (true) or Unchecked (false)

Examples

Example 1 — Array of objects with header row:

// Input
[
  {
    "header1": "row1-1",
    "header2": "row1-2"
  },
  {
    "header1": "row2-1",
    "header2": "row2-2"
  }
]

// Output
{
  "data": "aGVhZGVyMSxoZWFkZXIyCnJvdzEtMSxyb3cxLTIKcm93Mi0xLHJvdzItMg=="
}

Convert Text/JSON String to TXT (Base64)

Input Format

Accepts any input — plain text or JSON string
If input is valid JSON, it will be pretty-printed
If input is not valid JSON, it is kept as-is (lenient parsing)

Note: The TXT operation never throws an error on invalid JSON. Non-parseable input is treated as literal text.

Examples

Example 1 — Plain text:

// Input
Hello World

// Output
{
  "data": "SGVsbG8gV29ybGQ="
}

// Display in .txt:
Hello World

Example 2 — JSON string:

// Input
{"name":"Alice","age":30}

// Output
{
  "data": "ewogICJuYW1lIjogIkFsaWNlIiwKICAiYWdlIjogMzAKfQ=="
}

// Display in .txt:
{
  "name": "Alice",
  "age": 30
}

Example 3 — Non-JSON text:

// Input
"This is a report summary with {{{ invalid template"

// Output
{
  "data": "VGhpcyBpcyBhIHJlcG9ydCBzdW1tYXJ5IHdpdGgge3t7IGludmFsaWQgdGVtcGxhdGU="
}

// Display in .txt:
This is a report summary with {{{ invalid template

Convert Text/JSON String to DOCX (Base64)

Input Format

Accepts plain text, flat objects, arrays of objects (flat), or objects with arrays
Nested objects within a top-level object are not supported and will throw an error
Array items must only contain simple values (string, number, boolean, null) — nested objects or arrays within items are not supported
If input is not valid JSON, it is treated as plain text (lenient parsing)

Note: The DOCX operation never throws an error on invalid JSON. Non-parseable input is treated as literal text.

Invalid input:

{"project": "OMS", "metadata": {"owner": "Team A", "priority": "high"}}  // ❌ Nested object not allowed

[{"name": "John", "address": {"city": "NY"}}]  // ❌ Nested object in array item not allowed

[{"name": "John", "tags": ["a", "b"]}]  // ❌ Nested array in array item not allowed

Examples

Example 1 — Plain text:

// Input
Hello World

// Output
{
  "data": "<base64-encoded .docx binary data>"
}

// Display in .docx:
Hello World

Example 2 — Array of objects:

// Input
[{"name": "Alice", "age": 30}, {"name": "Bob", "age": 25}]

// Output
{
  "data": "<base64-encoded .docx binary data>"
}

// Display in .docx:
┌──────────┬──────────┐
│ name     │ age      │
├──────────┼──────────┤
│ Alice    │ 30       │
│ Bob      │ 25       │
└──────────┴──────────┘

Example 3 — Flat object:

// Input
{"project": "OMS", "version": "1.0"}

// Output
{
  "data": "<base64-encoded .docx binary data>"
}

// Display in .docx:
project: OMS
version: 1.0

Example 4 — Object with arrays:

// Input
{"project": "OMS", "modules": [{"name": "shared", "version": "1.0.0"}, {"name": "frontend", "version": "2.0.0"}]}

// Output
{
  "data": "<base64-encoded .docx binary data>"
}

// Display in .docx:
project: OMS
modules:
┌──────────┬──────────┐
│ name     │ version  │
├──────────┼──────────┤
│ shared   │ 1.0.0    │
│ frontend │ 2.0.0    │
└──────────┴──────────┘

Example 5 — Object with simple array:

// Input
{"name": "Report", "tags": ["urgent", "review", "approved"]}

// Output
{
  "data": "<base64-encoded .docx binary data>"
}

// Display in .docx:
name: Report
tags: urgent, review, approved

When to Use It?

Use Convert Node when:

You want to export data in standard formats (JSON, CSV, TXT, DOCX) for downstream processing.
You are integrating with systems that require base64-encoded file data.
You need to generate plain text reports from workflow data (TXT).
You need to generate formatted Word documents with tables from structured data (DOCX).

CFT (Trigger)

Crypto