JSON to Zod Converter

About the JSON to Zod converter

A TypeScript interface disappears at runtime, so it cannot tell you that an API returned something unexpected — it only promises that it will not. A Zod schema checks the data as it arrives and gives you the type for free through `z.infer`.

Paste a response and get both: a schema you can parse with, and a type derived from it that cannot drift out of sync.

Why validate at the boundary

Anywhere data enters your program from outside — an HTTP response, a config file, `localStorage` — a declared type is an assumption rather than a fact. Parsing with a schema turns a confusing failure deep in your code into a clear one at the edge.

Because the type comes from `z.infer`, there is only one definition to maintain instead of a type and a validator that can disagree.

What to tighten by hand

The generated schema is structural: it knows a field is a string, not that it is an email or a UUID. Adding `.email()`, `.uuid()`, `.min()` and `.optional()` is where the real value is, and only you know which apply.

Integers are emitted as `z.number().int()` where the sample had no fractional part — check that this is genuinely a constraint and not a coincidence of the sample.

Types come from one sample

Everything here is inferred from the JSON you paste. A field that happens to be absent from your example will be absent from the output, and a field that is null in the sample cannot be typed at all — only marked as nullable.

Paste the richest response you have, then review the optional and nullable fields by hand. Treat the result as a fast first draft rather than a schema.

Frequently asked questions