Expand / collapse
Collapse unrelated branches and keep the path you care about. A few shallow levels expand by default; click deeper nodes as needed.
Objects show field counts and arrays show item counts, which helps estimate size.
You can also load data via #data={"name":"Ada"} or #url=... It is cleared from the address bar after it is read.
Show JSON as a collapsible tree. When nesting is deep and fields are many, that is clearer than hunting for brackets in plain text, and you can see each node’s type. Click a node to copy its JSONPath for tests or the JSONPath page.
Collapse unrelated branches and keep the path you care about. A few shallow levels expand by default; click deeper nodes as needed.
Objects show field counts and arrays show item counts, which helps estimate size.
Objects, arrays, strings, numbers, booleans, and null are distinguished. Avoid treating "123" and 123 as the same thing.
Useful during integration to confirm the types the backend actually returns.
The tree shows structure; use the field-detection page for timestamps and colors. Locate a path on the tree, then check whether a leaf is a special format.
Read-only browsing. You do not edit business data on the tree.
{
"project": "json-toolbox",
"config": {
"theme": "light",
"limits": { "maxSizeMB": 10 }
}
}
the root is an object → config is an object → limits is an object → maxSizeMB is a number. Click nodes instead of counting brackets.
This page is mainly for viewing. After edits, paste back into the formatter to validate.
Extract a subtree with JSONPath first, then expand it.
It tries to copy that node’s JSONPath (for example $.user.id ) to the clipboard, which is handy for assertions.
Start with valid JSON. Invalid text cannot be drawn as a tree—validate or format first.
The tree is for “seeing structure.” JSONPath is for “taking values in bulk.” For large documents, using both is faster than expanding everything.