JSON Schema Faker

JSON Schema Faker turns JSON Schema definitions into synthetic JSON for developers who need schema-conforming fixtures or API data. It is best suited to Node.js, Bun, and browser workflows built around JSON Schema. The seed and reference controls make repeatable generation practical, but synchronous I/O limits rule it out for schemas that need asynchronous resolution.

Overview

This is a focused generator, not a hosted data service or a relational-data tool. It supports JSON Schema 2019-09 and 2020-12, and the project identifies its license as MIT. The project describes its production dependencies as zero, keeping the core lean; optional faker.js and chance.js integrations, along with YAML output, require extensions and packages.

Key features

Repeatable generation and CLI

A numeric seed makes output deterministic: use the same seed again and the generator returns the same results. That is valuable for stable test fixtures and debugging, because the sample data does not change between runs. The jsf command can generate JSON from schema files, write it to a file, and produce multiple items as an array, making it useful in scripts and batch fixture workflows.

References, formats, and customization

Reference handling includes cycle detection, configurable recursion depth, and remote reference resolution. This supports schemas that reuse definitions or include recursive structures. Built-in generators cover common formats including email, date-time, URI, IP addresses, and UUID; regex patterns are handled with an AST-based string generator. Users can also register custom formats and define generators for custom keywords, which helps adapt generation to domain-specific schemas.

Generation limits

Synchronous generation works only when the schema can be resolved without I/O. An asynchronous resolver or extension makes it throw, so workflows relying on asynchronous resolution need a different approach. JSON output is central; YAML output is possible through an extension that requires the YAML package.

Pricing

JSON Schema Faker costs nothing under the MIT License, with npm and Bun installation. There are no paid tiers to weigh against the free option. This suits teams comfortable installing and running an open-source package themselves; it does not provide a hosted generation service or a relational-data model.

Platforms

The project runs in Node.js and browsers and is built for Bun and Node.js. It is categorized for Linux, macOS, Windows, web, and self-hosted use. Deployment is self-hosted, so the generator fits environments where teams run it within their own applications or workflows.

Who it's for

Choose JSON Schema Faker when tests or API workflows need synthetic data shaped by JSON Schema, especially when repeatability, reference handling, and custom generators matter. It is less suitable when data must represent relationships across tables, or when schema resolution depends on asynchronous I/O.

Pros and cons

Pro: Seeded deterministic output makes fixtures repeatable, which helps keep tests and debugging consistent.

Pro: Reference controls and common format generators support schemas with reused definitions and structured values.

Pro: Custom formats and keyword generators allow teams to extend behavior for their own schemas.

Con: Synchronous generation throws when an asynchronous resolver or extension is involved, restricting I/O-dependent schemas.

Con: It does not generate relational data, so it is not a fit for connected-table test datasets.

Con: YAML output and faker.js or chance.js integrations require extensions rather than being part of the zero-production-dependency core.

Alternatives

For a broader set of test-data options, see Test Data Generation Tools.

dbForge Studio for PostgreSQL is worth considering when a PostgreSQL-focused tool and visual data editing matter more than schema-driven JSON generation; its Express plan includes a Visual Data Editor and data import and export.

Benerator is an open-source alternative for readers who need generation and anonymization; its Community Edition is single-threaded.

Mockaroo may suit users who prefer browser-based generation or need an API, with its free plan capped at 1,000 rows per file and 200 API requests per day.

YData SDK is an alternative for readers who need connectors, with 20+ connectors included in its free plan.

Bogus is an option for teams building on.NET, distributed as a free open-source package.

Redgate SQL Provision is a paid option for provisioning or masking production data, with capacity tiers based on total production data volume.

MOSTLY AI is another freemium option for readers considering synthetic-data tools.

Snowfakery is a free open-source alternative with no paid plans or usage limits stated.

Verdict

JSON Schema Faker is a strong choice for developers who want free, self-hosted, repeatable synthetic JSON aligned to modern JSON Schema versions. Its useful combination of seeded output, reference handling, and extensibility makes it especially practical for schema-led fixtures. Look elsewhere if you need relational datasets or asynchronous schema resolution.