JSON Exploration Workbench: Explore and Redact JSON
September 23, 2026

JSON Exploration Workbench is a local app for making sense of a large JSON response. Paste it in, filter down to the nested paths you care about, see the types and a preview of each value, and redact sensitive fields before you share it — then export the cleaned JSON and a short record of what you did. It's inspired by fx, the popular terminal JSON viewer.
| Type | Desktop app |
| Category | Development · Data & analytics · Productivity |
| What's inside | 1 app · runs locally · no code execution, no network |
| Inspired by | antonmedv/fx on GitHub |
| Cost to use | Free |
What it does
An API response or a config export can run to thousands of lines. The questions you actually have are narrow: where's the field I need, what type is it, and is anything in here I shouldn't paste into a ticket?
Filter nested paths. Narrow a large document down to the part you need instead of scrolling through all of it.
Inspect types with bounded previews. See what each value is — object, array, string, number — with a short preview, so a long string or a huge array doesn't flood the screen.
Surface structural risks. It points out arrays that mix different kinds of items and structures nested unusually deep — the two shapes most likely to break code that reads the data.
Redact sensitive values. Values under sensitive-looking keys are masked in a deterministic way, so the same input always produces the same redacted output. That's what you want before pasting a real response into a bug report or a chat.
Export the result. Save the transformed JSON, plus a Markdown note describing what you filtered and redacted.
What it won't do
- open files from your disk or fetch URLs — you paste the JSON in
- run code or evaluate expressions
- make network requests
- guarantee that redaction caught everything
Redaction by key name is a safety net, not a guarantee. A secret stored under an innocent-looking key will pass straight through, so read the output before you share it.
When to reach for it
- You're debugging an API response and need one field from a large payload
- You're about to paste a real response into a ticket, a chat, or a support request
- You want to understand the shape of an unfamiliar JSON export
- A parser keeps failing and you suspect a mixed array or a deeply nested value
- You'd rather not paste production data into an online JSON viewer
How it differs from fx
fx is a much bigger tool, built for the terminal.
| fx (original) | JSON Exploration Workbench (Taku) | |
|---|---|---|
| Runs in | Your terminal, as an interactive viewer and command-line processor | A local app in Taku |
| Input | Files, pipes, and streams, including JSON per line | JSON you paste in |
| Transforming data | JavaScript expressions on the command line | Path filters, with no code evaluation |
| Formats | JSON, plus YAML and TOML | JSON |
| Redaction | Not a built-in focus | Deterministic masking of sensitive keys |
If you live in a terminal, fx — or jq, the other standard tool — is the right choice. The Workbench is for exploring and cleaning a JSON snapshot without installing a command-line tool or running any code.
What's actually inside
One app. It works only on the JSON you paste, with no file access, no code execution, and no network requests.
Setting it up
The GitHub route (the original). fx installs with a package manager — Homebrew, Snap, Scoop, pacman, pkg, or Nix — or with go install, or in Docker, per its installation page. Then you pipe JSON into it in a terminal, like curl ... | fx.
The Taku route. Open JSON Exploration Workbench in Taku and paste your JSON.
Who made it
fx is an open-source project by Anton Medvedev, released under the MIT license. JSON Exploration Workbench is an independent adaptation listed in the Taku marketplace. It isn't an official fx release.
If you don't live in a terminal
You don't need to understand GitHub, Codex, or Claude Code to look inside a JSON file. Terminal tools like fx and jq are excellent, and they assume you're comfortable installing command-line tools and piping data between them.
Taku is a desktop workspace built for people who'd rather skip that: find something someone already built, open it in your own workspace, and use it. Browse the free app library for more developer tools, or read about data automation tools for what happens once you know what's in the data. If you're designing the database the data will land in, ChartDB Local Schema Designer drafts the tables and relationships.
Download Taku and explore your next payload safely. Taku is in Beta, and the Mac app is available now; there's also an experimental, unsigned Windows build for Windows 10 and 11.
FAQ
What does JSON Exploration Workbench do?
It lets you paste JSON, filter nested paths, inspect types with short previews, spot mixed arrays and deep nesting, redact sensitive values, and export the result with a Markdown note.
Does it send my JSON anywhere?
No. It makes no network requests and works only on what you paste.
Can I use JavaScript expressions like in fx?
No. It deliberately doesn't evaluate code. Use fx or jq in a terminal if you need expressions.
Is the redaction safe to rely on?
Treat it as a first pass. It masks values under sensitive-looking keys, but it can't know that an innocently named field holds a secret, so check the output.
Is it free?
The app is free in the marketplace. For what Taku itself costs, see pricing.
Key points
- Explores pasted JSON: filter paths, inspect types, preview values.
- Flags mixed arrays and deep nesting, the shapes that break parsers.
- Deterministically redacts sensitive keys before you share a payload — but check the result.
- Local only: no file access, no code evaluation, no network.
- Inspired by fx, the terminal JSON viewer; use fx or jq if you want expressions.