JSON to Zod Schema Converter

What is a Zod schema?

Examples

{"id":101,"name":"Ada Lovelace","email":"ada@example.com","active":true,"roles":["admin","user"]}

import { z } from "zod";

export const RootSchema = z.object({
  id: z.number().int(),
  name: z.string(),
  email: z.string(),
  active: z.boolean(),
  roles: z.array(z.string()),
});

export type Root = z.infer<typeof RootSchema>;

{"repo":"zod","owner":{"login":"colinhacks","id":100}}

import { z } from "zod";

export const OwnerSchema = z.object({
  login: z.string(),
  id: z.number().int(),
});

export const RootSchema = z.object({
  repo: z.string(),
  owner: OwnerSchema,
});

export type Root = z.infer<typeof RootSchema>;

{"users":[{"id":1,"nick":"x"},{"id":2}]}

import { z } from "zod";

export const UserSchema = z.object({
  id: z.number().int(),
  nick: z.string().optional(),
});

export const RootSchema = z.object({
  users: z.array(UserSchema),
});

export type Root = z.infer<typeof RootSchema>;

{"items":[{"sku":"A1","discount":0.1},{"sku":"A2","discount":null}]}

import { z } from "zod";

export const ItemSchema = z.object({
  sku: z.string(),
  discount: z.number().nullable(),
});

export const RootSchema = z.object({
  items: z.array(ItemSchema),
});

export type Root = z.infer<typeof RootSchema>;

How to convert JSON to Zod

1. 1

   Paste your JSON

   Drop a JSON object, array, or API response into the input box. Conversion starts instantly.

2. 2

   Tune the output

   Rename the root schema and toggle the import line, z.infer type, or export keyword to match your project's style.

3. 3

   Copy or download

   Grab the generated Zod schema with one click and paste it straight into your validation layer.

Common Use Cases

Validate API responses

Turn a sample REST or GraphQL response into a Zod schema and parse it at the fetch boundary, so malformed data fails fast with a clear error instead of leaking through your app.

Type-safe forms and inputs

Generate a schema for form or query input and reuse the z.infer type across your UI and handlers for end-to-end type safety.

tRPC and AI tool schemas

Bootstrap input and output schemas for tRPC procedures or AI structured-output tools from an example payload instead of writing them by hand.

How the conversion works

Structural inference

Each object becomes a named schema; identical shapes are declared once and reused. Arrays of objects merge key by key, and keys absent from some items become .optional().

Correct type mapping

Integers map to z.number().int() and decimals to z.number(). Strings, booleans, and nulls map to z.string(), z.boolean(), and z.null(); an array of mixed primitives becomes a z.union().

TypeScript types included

A z.infer type alias is emitted for the root schema, so the schema stays the single source of truth and your types never drift from your validation.

100% client-side

Parsing and generation run in your browser with no network calls, so your data stays private.

Tips for clean Zod schemas

Name your root schema

Set a meaningful root name such as User or ApiResponse instead of the default Root for readable, self-documenting code.

Refine z.unknown() and unions

Empty or mixed arrays produce z.unknown() or a broad z.union(). Paste a richer, representative sample so the tool infers precise types, then tighten anything that stays ambiguous.

Validate at the boundary

Call safeParse() where untrusted data enters — API responses, form input, webhooks — so bad data is caught before it reaches your application logic.

Frequently asked questions