Free JSON viewer online: what to look for and how to use one
What a good free online JSON viewer should do, how to check one is safe for your data, and how to use Cards, Table, and Tree views.
There are plenty of free JSON viewers on the web, and they are not all the same. Some only pretty-print text. Some choke on a moderately large document, and some send what you paste to a server you know nothing about. If you only need a quick look at a payload, it is still worth thirty seconds to check what a tool actually does. This guide lists the things to look for, explains the privacy question in plain terms, and then shows how a viewer fits into an ordinary debugging session.
What a free viewer should do at minimum
A useful viewer has four basic abilities. It accepts text by paste or file drop. It validates, meaning it tells you whether the input is well-formed under the JSON grammar and, if not, where the first problem is. It formats, producing consistent indentation. And it offers at least one structured view, such as a tree, so you are not reading raw text.
Beyond that, the useful extras are the ones that match how your data is shaped. A table view helps with arrays of records. A collapsible tree helps with deep nesting. The ability to edit a value and copy the result back out saves a round trip to your editor. Anything else is a bonus.
The privacy question
The question to ask about any online tool is where your text goes when you paste it. Some tools process everything in the browser, so the data stays on your machine. Others post it to a server for parsing or sharing. Neither approach is automatically wrong, but you should know which one you are using, especially for internal data.
View My JSON is a free web app at app.viewmyjson.com whose core viewing and editing run in the browser. Pasted text is parsed locally, and you do not need an account for the core tools. Even so, a sensible rule applies to every online tool: remove secrets before you paste. API keys, bearer tokens, session identifiers, and customer details do not belong in any web page you do not control, and your employer may have rules about it.
Using a viewer on a real payload
Say you have called an endpoint that lists deployments and one of them is behaving oddly. Copy the response body, paste it into the viewer, and validate. If it parses, format it and read the top-level structure. The response below is a plausible example.
The top-level object has a count and an items array. Each item has a name, a status, and a nested replicas object. You can see the shape before reading a single value, and you can decide whether Table, Cards, or Tree is the right next step.
{
"count": 3,
"items": [
{ "name": "api", "status": "running", "replicas": { "ready": 3, "desired": 3 } },
{ "name": "worker", "status": "degraded", "replicas": { "ready": 1, "desired": 4 } },
{ "name": "cron", "status": "running", "replicas": { "ready": 1, "desired": 1 } }
]
}Which view for which job
For the deployments example, the items array is a list of similar records, so a table works well. Name, status, and the replica counts line up in columns, and the degraded worker, with one of four replicas ready, is easy to find. Cards would also work if you prefer each deployment as a small block with labelled fields. A tree is more useful if the replicas object were much larger or contained several levels of its own.
All three, along with Format and Validate, are free in View My JSON. The Pro plan adds Charts, 3D, Compare, and Rich view, which you may reach for when you want to graph a numeric series or put two versions of a document side by side. You can ignore them entirely if you only want to read data.
Reading a validation error
When validation fails, the message usually includes a line and column, or a character position, and a description such as unexpected token or expected a comma. The position is where the parser noticed the problem, which is sometimes just after the real mistake. A missing comma between two properties is typically reported at the start of the second one.
Here is a short example of the most common failure: a trailing comma. It is accepted by many programming languages and by some tools that parse JSON-like formats, but RFC 8259 does not allow it.
{
"name": "demo",
"tags": ["a", "b",],
}Limits worth knowing about
A browser tab has finite memory, so very large documents may be uncomfortable in any browser-based viewer, and this is true of all of them. If you routinely deal with files that are hundreds of megabytes, a command-line tool or a script is a better place to start, perhaps to extract the part you care about before you open it visually.
A viewer also cannot repair meaning. If a field contains the wrong value, or a date is in the wrong format, the tool will display it faithfully. Its job is to make the problem visible, not to decide what is correct.
Checking a viewer with awkward input
A quick test tells you more than a feature list. Paste a document with the sort of values that trip up weaker tools, and see whether each one survives intact. The sample below contains an empty object, an empty array, a null, a non-ASCII string, an escaped quotation mark, and a number written in exponent form. All of it is valid JSON.
A good viewer should show the empty containers as empty rather than hiding them, keep null distinct from the string "null", display the accented character correctly, and not alter the number when you copy the document back out. If any of that goes wrong, treat the tool with caution for important data.
{
"empty": {},
"none": [],
"missing": null,
"city": "Zürich",
"quote": "She said \"ready\"",
"big": 1.5e3
}A fair way to choose
Try a viewer with a document you already understand and see whether the output matches what you expect. Paste something with a null, an empty array, a nested object, a Unicode character, and a long number, and look at how each is displayed. A tool that shows all of those clearly is likely to be reliable on documents you do not yet understand.