JSON

JSON Schema Validator

Paste or upload an instance and schema, then inspect every reported validation issue without sending either document to a server.

Loading local tool…

How this tool works

JSON Schema describes constraints on a JSON instance. A schema can distinguish required from optional properties, restrict primitive types, define nested object and array rules, and compose reusable constraints.

This page first parses both editors as JSON, then validates the schema itself, then validates the instance. Keeping those phases separate prevents a malformed schema from being misreported as bad instance data.

Examples

  • Require an orderId string and a positive quantity.
  • Report nested array item failures with instance and schema paths.
  • Reject a value outside an enum without converting it.

Common use cases

API payload contract checks
Fixture review
Negative test design

How to read the report

Instance paths identify the value that failed, while schema paths identify the constraint that produced the finding. The keyword explains the class of rule, such as required, type, enum, minimum, or additionalProperties.

A missing required property is normally attached to its containing object because no child value exists to point at. Array failures use numeric item positions.

Important boundaries

  • Schema validity is different from application or business validity.
  • Format handling follows Ajv defaults and this page does not add network-backed format checks.
  • Remote references are not fetched; bundle reusable definitions into the pasted schema.
  • Large documents are capped at 2 MiB per editor to limit browser memory work.

QA scenarios

  • Run the smallest valid instance first.
  • Omit each required property independently.
  • Exercise null, wrong-type, boundary, enum, nested-array, and additional-property cases.
  • Export a report with the schema revision used by the test evidence.

Invalid instance

Invalid/failure example

{"quantity":0}

Expected: required: orderId; minimum: quantity must be >= 1

Not a business-rule engine

A passing schema result does not prove authorization, database state, cross-field business rules, or deployed API behavior.

Related guides

Official references

FAQ

Does this tool send the pasted data or execute an API request?

No. Analysis runs in this browser tab. The tool does not call pasted URLs, resolve remote references, or submit the input to a DevPouch processing endpoint.

Which JSON Schema dialect is supported?

The validator uses Ajv 8 in its Draft 2020-12 mode. It validates schemas before instances, reports all discovered errors, and does not coerce input values.

Does valid JSON mean the instance matches the schema?

No. JSON syntax validation only proves the text parses. Schema validation applies types, required properties, bounds, enums, applicators, and other declared constraints.

Related tools