JSON Schema Examples: 8 Patterns You Can Copy

A basic object with required fields

Every schema starts with a type. For an object, you list its properties and which ones are required: type set to object, a properties map describing each field, and a required array naming the must-have keys. Validation fails if a required key is missing or if a value has the wrong type. This single pattern covers the majority of everyday validation — confirming a payload has the fields your code expects.

{
  "type": "object",
  "properties": {
    "id":   { "type": "integer" },
    "name": { "type": "string" }
  },
  "required": ["id", "name"]
}

Types and the type keyword

• string, number, integer, boolean, object, array, and null are the available types.
• A property can allow multiple types with an array, such as type: [string, null] for a nullable field.
• integer is stricter than number — it rejects values with a fractional part.
• additionalProperties: false rejects any key not listed in properties, locking the object down.

String formats and patterns

Strings can be constrained beyond just being text. The format keyword validates well-known shapes like email, date-time, uri, and uuid. For custom rules, pattern applies a regular expression — for example requiring a value to match a phone or product-code format. minLength and maxLength bound the length. These turn a loose string field into a precise contract.

{
  "type": "string",
  "format": "email",
  "minLength": 5
}

Numbers, ranges, and enums

Numeric fields use minimum and maximum (and the exclusive variants) to bound a range, and multipleOf to require, say, multiples of 5. For a fixed set of allowed values, the enum keyword lists them: a status field might be constrained to the enum active, archived, and deleted, rejecting anything else. Enums are the cleanest way to validate a closed set of options.

{
  "status": { "enum": ["active", "archived", "deleted"] },
  "age":    { "type": "integer", "minimum": 0, "maximum": 120 }
}

Nested objects and arrays

Schemas nest just like the data. A property whose type is object gets its own properties and required list, validating sub-structures to any depth. For arrays, the items keyword describes what each element must look like — for instance an array of objects each requiring an id. minItems, maxItems, and uniqueItems add list-level constraints like non-empty and no duplicates.

{
  "type": "array",
  "minItems": 1,
  "items": {
    "type": "object",
    "required": ["id"],
    "properties": { "id": { "type": "integer" } }
  }
}

Conditional validation

Advanced schemas express rules like 'if type is business, then a tax ID is required' using if/then/else, or combine sub-schemas with allOf, anyOf, and oneOf. These let one schema validate several related shapes — for example a payment object that requires different fields depending on its method. Reach for them when a flat list of properties cannot capture the real rules.

Generating and testing schemas

You do not have to write schemas by hand. A JSON Schema Generator infers a starting schema from a sample document, which you then tighten with formats, ranges, and required fields. A JSON Schema Validator checks real payloads against your schema and reports exactly which rule failed, so you can iterate quickly toward a contract that matches your data.