Recognise the basic building blocks
An object is wrapped in curly braces and contains key–value pairs. An array is wrapped in square brackets and contains an ordered list of values. JSON values can be strings, numbers, booleans, null, objects or arrays. Property names and string values use double quotes.
JSON is not exactly the same as a JavaScript object literal. Standard JSON does not permit comments, single-quoted strings, functions, undefined values or trailing commas. A snippet can look familiar and still be invalid JSON.
{
"name": "MODEL AI",
"tools": 35,
"freeTier": true,
"categories": ["study", "writing", "images"],
"ownerNote": null
}Common syntax errors
Missing commas between properties, extra commas after the last item, mismatched braces and unescaped double quotes are common problems. A string containing a quotation mark must escape it with a backslash. Newline characters inside a JSON string also need escaping rather than being entered as a raw line break.
When a validator reports an error location, inspect the line just before the reported position too. A missing comma earlier in the file can make the parser complain later than the original mistake.
{
"title": "A story",
"ready": true
}Formatting does not validate the application rules
A formatter can indent JSON and make nested objects easier to inspect, but valid JSON may still contain an impossible date, a misspelled field name or a number outside an application’s allowed range. Syntax validation asks whether the text follows JSON grammar; schema or business validation asks whether the structure and values are appropriate for a particular purpose.
For instance, {"age": -7} is syntactically valid JSON, but an application may reject a negative age. Likewise, a product record can be valid JSON while missing a field the application needs. Check both layers where the data will be used.
Handle large or sensitive files carefully
If a JSON document is very large, work on a copy and use a formatter with line numbers. Avoid pasting secrets, access tokens or customer records into online formatters unless you understand the processing path. A local browser formatter is preferable for sensitive content when it genuinely runs locally.
After formatting, compare the output structure and test it with the application or a small representative sample. Formatting should change whitespace, not the meaning of the data. If the file is important, keep the original and use version control or a separate backup.
A safe debugging workflow
First, read the error message and identify whether the issue is parsing, schema validation or application behaviour. Second, reduce the data to a small example that still fails. Third, validate the small sample and fix one syntax issue at a time. Fourth, restore the full file and run the application’s own checks. This keeps debugging focused.
Do not delete an unfamiliar field just because a parser points nearby. A field may be valid but use the wrong type or be required by another part of the system. Confirm what the receiving application expects before changing the data model.
When to use minified JSON
Pretty-printed JSON is easier for people to read and review. Minified JSON removes extra whitespace and can reduce file size slightly, which may be useful for machine transport. Keep the readable source for development and review, and produce minified output only when a target system expects it or it helps the workflow.
Never treat minification as a security feature. It does not encrypt data or remove secrets. If credentials appear in a JSON file, remove or rotate them appropriately instead of hiding them with formatting.