Skip to content

9paradox/json-query-flow

Repository files navigation

Json Query Flow

Json Query Flow is a tool for exploring, querying, and transforming JSON data using a visual, node-based workflow. It is intended for developers and data analysts who work with complex or unfamiliar JSON structures and need a fast, interactive way to understand and transform data.


Overview

Json Query Flow provides a canvas where JSON data, queries, and outputs are connected as nodes. It focuses on exploration, validation, and prototyping rather than production execution.

The tool helps users reason about JSON shape, experiment with queries, and verify results before those queries are moved into application code, scripts, or data pipelines.


Screenshot

You can try the tool here:
https://json-query-flow.9paradox.in/

Json Query Flow – visual JSON query and transformation tool


What this tool is

Json Query Flow allows you to:

  • Load and inspect raw JSON data
  • Write and test JSONata expressions interactively
  • Chain transformations using a visual flow
  • Inspect outputs as JSON, tables, or charts
  • Optionally use AI assistance to help draft queries

It runs primarily in the browser and is designed for iterative exploration.


Why this tool exists

Working with JSON often involves friction:

  • API responses are deeply nested or undocumented
  • Query logic is written blindly and tested later
  • Data analysts need to explore structure before analysis
  • Visualization requires extra scripts or tools

Json Query Flow reduces this friction by providing immediate feedback and visual context, making it easier to understand data and validate transformations early.


Who this tool is for

  • Backend engineers working with API responses
  • Frontend engineers transforming data for UI consumption
  • Data analysts exploring JSON exports or API data
  • Developers prototyping transformation logic
  • Teams debugging or validating JSON-based integrations

This tool is not intended to replace production data pipelines or analytics platforms.


How the tool works

Main Data Node

  • Holds the input JSON
  • Acts as the source for downstream nodes

Query Nodes

  • Accept JSON from connected nodes
  • Support JSONata and natural language queries
  • Emit transformed results

Visualization

  • Outputs can be viewed as:

    • Raw JSON
    • Table (for arrays of objects)
    • Bar and pie charts with aggregation options

Execution model

  • Queries execute client-side where possible
  • AI-assisted queries are routed through a Cloudflare Worker
  • API keys are forwarded per request and not persisted by the application

AI integration (optional)

AI features are optional and require a user-provided API key.

  • The key is sent securely to a Cloudflare Worker
  • The worker forwards requests to the selected AI provider
  • The application does not store or log API keys

AI assistance is intended to help with exploration and query drafting, not to replace manual validation.


Built with

  • JSONata – JSON query and transformation language
  • React Flow – Node-based editor and canvas
  • Monaco Editor – JSON and query editing
  • shadcn/ui – UI components and accessibility primitives
  • Cloudflare Workers – Secure edge runtime for AI requests

Feedback and contributions

Feedback, bug reports, and contributions are welcome.

  • Use GitHub Issues to report bugs or unexpected behavior
  • Provide reproducible examples where possible
  • Pull requests are welcome for fixes, improvements, or documentation updates

This project is evolving and open to practical improvements.


License

MIT License

About

A visual JSON exploration and query builder with support for multiple query languages and visualizations.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors