Structured Output Validator
Check structured JSON against a schema.
About this tool
Use it to check structured output or tool-call arguments before your code trusts them. Paste the JSON and the schema, and the validator lists missing required fields, wrong types and values outside an enum. It works entirely in your browser.
How to use
How it works
The check covers type, required, properties, items and enum, recursing through nested objects and arrays and reporting paths like $.items[2].qty. A clean pass returns valid true with an empty error list and the parsed data. It does not enforce numeric ranges, string patterns or combinators.
Worked example
A valid payload
- Input
- JSON: {"name":"Ada","email":"a@b.c"} Schema: {"type":"object","required":["name","email"],"properties":{"name":{"type":"string"}}}
- Output
- { "valid": true, "errors": [], "parsed": { "name": "Ada", "email": "a@b.c" } }
- Walkthrough
- Both required keys are present and name is a string, so the check passes. Remove email and it fails with $.email: required.
Limitations
- Only type, required, properties, items and enum are enforced; other schema keywords are ignored.
- Properties not listed in the schema are allowed.
- It does not coerce types ("30" is not accepted as an integer).
Questions
How do I validate a function call's arguments?
Does it check for extra properties?
Does it work with OpenAI or Anthropic structured outputs?
Is it the same as the JSON output validator?
Features
- Nested path errors
- valid flag and parsed data
- Runs in your browser
Related resources
Guides and videos from around the web—opens on YouTube or the publisher’s site.
TensorFlow 2.0 Complete Course
freeCodeCamp.org
Practical ML stack context next to AI utilities.
Watch on YouTubeFinding mean, median, and mode
Khan Academy
Baseline stats for accuracy and evaluation helpers.
Watch on YouTube
- Metrics and scoring
scikit-learn
Precision, recall, F1 definitions used by AI metric tools.
- OpenAI Research
OpenAI
Research notes for people exploring AI tooling trends.