JSON Validator
Check any JSON for syntax errors with the exact failure position, plus a structure summary: objects, arrays, keys, leaf values and nesting depth.
Reviewed 17 November 2026 · Free · No account needed
How to use
- Paste the JSON to validate.
- Read the verdict and, for errors, the exact line/column position.
- Valid JSON is re-formatted with 2-space indent for inspection.
Example
Input: {"a": 1, "b": [2, 3}
Output: Invalid — unterminated array, reported near the exact line/column
How it works
The input is parsed as JSON. Valid input is confirmed; invalid input reports the parser's error and the position where parsing failed.
The input is parsed with JSON.parse. Valid documents are re-printed with 2-space indent and summarised — objects, arrays, keys, leaf values and maximum nesting depth are counted in one walk of the tree. Invalid input reports the parser's error with the failing position converted to a line and column.
{"a": 1, "b": [2, 3} is invalid — the unterminated array is reported near its exact line and column. A valid document instead shows, for example, 1 object · 1 array · 2 keys · max depth 2.
Edge cases to know
- Validation follows JSON.parse exactly (RFC 8259): trailing commas, comments and single quotes are errors.
- Duplicate keys keep the last value, matching browser behaviour.
- Multi-megabyte documents may be slow on low-memory devices.
Limits & compatibility
- Validation follows JSON.parse exactly (RFC 8259) — trailing commas are errors.
- Duplicate keys keep the last value, matching browser behaviour.
- Very large documents (multi-MB) may be slow on low-memory devices.
Privacy
Your JSON never leaves the browser. There is no server call, no storage and no logging of content.
Guides that use JSON Validator
Related tools
Questions about JSON Validator
What rules does the validation follow?
Exactly what JSON.parse accepts, which is the RFC 8259 grammar: no comments, no trailing commas, no single-quoted strings. When parsing fails, the error is reported with the line and column position of the failure, so a stray comma in a long document takes seconds to find rather than minutes.
What does the structure summary tell me?
For valid input it counts objects, arrays, keys and each kind of leaf value — strings, numbers, booleans and nulls — plus the maximum nesting depth. It is a quick shape check before you wire a payload into code: an unexpectedly deep document or a key count that looks wrong usually means the wrong export was pasted.
Why is my JSON rejected when another tool accepts it?
Most often trailing commas — many editors and linters tolerate them, but JSON.parse treats them as syntax errors, and so does this validator. Duplicate keys are the opposite case: they parse fine, and the last value wins, matching browser behaviour. Everything runs in your browser, so nothing about the document is uploaded while you debug it.
Is there a size limit?
No hard cap — but validation, the structure summary and the re-formatted 2-space output all run in your browser tab, so multi-megabyte documents may be slow on low-memory devices. If a huge payload stalls, validate a smaller slice first; the syntax rules are the same in every chunk.