JSON Schema for Beginners: Validate Your Data with a Few Keywords

JSON Schema is a standard way to describe what valid JSON looks like: which fields exist, what type each one is, and what values are allowed. Once you have a schema you can validate incoming data, generate documentation, and catch mistakes before they reach your code.

A first schema

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "name": { "type": "string", "minLength": 1 },
    "age": { "type": "integer", "minimum": 0 },
    "role": { "enum": ["admin", "user"] },
    "tags": { "type": "array", "items": { "type": "string" } }
  },
  "required": ["name", "role"],
  "additionalProperties": false
}

The keywords you'll use most

• type: string, number, integer, boolean, object, array, or null.
• properties: describes the fields of an object; required lists which ones must be present.
• items: describes the elements of an array.
• enum: restricts a value to a fixed list; const requires an exact value.
• minimum and maximum for numbers; minLength, maxLength, and pattern for strings.
• additionalProperties: false rejects fields you didn't list, which catches typos.

Generate first, then refine

The quickest way to start is to paste a sample document into a generator that infers a schema, then tighten it by hand. A generator can only see your example, so it won't know which fields are optional or what ranges are valid.

1. 1Generate a schema from a representative sample.
2. 2Mark truly required fields in required.
3. 3Add constraints: ranges, lengths, enums, and formats.
4. 4Validate both good and deliberately bad samples to make sure the schema rejects what it should.

Good habits

• State the draft with $schema so tools know which rules to apply. Draft 2020-12 is the current one.
• Reuse repeated shapes with $defs and $ref instead of copy-pasting.
• Keep schemas in version control next to the code that depends on them.
• Remember that a schema validates structure, not business rules such as "end date after start date".