Does the JSON viewer upload my file?
No. The tree is built in the browser from the text you pasted.
Paste JSON and walk it as a collapsible tree. Expand only the branch you care about โ the document is parsed in this tab and never leaves the device.
JSON formatter
Run a tool to see a polished result here
Pretty-printed JSON is still a wall of text: every nested object costs another indent level on the screen. The viewer maps objects and arrays onto foldable nodes, auto-folds large arrays, and lets you change a value in place. When you need indented source to drop into a PR, switch to the beautifier; when you need to hunt `user.profile.roles[1]`, stay on the tree.
Need indented text to paste into a file? The JSON beautifier pretty-prints the same document without the tree chrome.
Already know the path you want? JSONPath extracts matches without clicking through nodes.
Blank tree? The document is probably invalid โ check it with the JSON validator first.
Looking at two responses rather than one tree? JSON compare reports added, removed, and changed paths.
A 100-item list is painful as pretty-printed text and painful as 100 open nodes. The viewer auto-folds long arrays so you expand a window of indexes at a time. That is the main reason this page exists alongside the beautifier: the beautifier must show every line, the tree does not.
Objects show keys. Arrays show indexes. Strings, numbers, booleans, and null are labeled as leaves so you can spot `1` vs `"1"` without reading quotes in a 4 kB dump. Duplicate keys that survived parsing appear as the last value; the status line still warns they existed.
If you already know the path (`$.store.book[*].author`), JSONPath is faster than expanding nodes. If you do not know the path, the tree is how you discover it. Many people use both: viewer to find the field, JSONPath to extract it into a list.
The sample nests a user profile, feature flags, and an items array so you can collapse flags, expand roles, and open a single SKU without pretty-printing the whole document. Swap in a production payload of your own; it stays in this tab.
{
"user": {
"id": "u_18",
"profile": { "name": "Ada", "roles": ["admin", "editor"] }
},
"flags": { "dark": true, "beta": false },
"items": [
{ "sku": "fmt", "qty": 2 },
{ "sku": "xlsx", "qty": 1 }
]
}Open the viewer when the problem is โwhat is inside this blob?โ rather than โmake it prettyโ or โturn it into Excelโ. It is the right URL for json viewer / json tree searches. It is the wrong URL for schema validation or comparing two files side by side.
Also try JSON beautifier, JSON validator, JSON compare, JSONPath, or JSON to Excel. The JSON formatter on the homepage is the general workspace.
No. The tree is built in the browser from the text you pasted.
Yes for inspection. Changing a leaf writes back into the document. Treat it as a scratchpad, not a full IDE.
The viewer needs a parseable document. Use the validator or Repair, then run the viewer again.
Pretty print is linear text. A tree lets you collapse 90% of a payload and still see one nested field.
No. It is a local tree for viewing and light edits. It does not load JSON Schema.