JSON Schema Validator

Overview

You send a JSON Schema and a document, and we tell you whether the document satisfies it. Every problem comes back with the exact place in your document it happened, the schema rule that rejected it, and a readable message, so you can fix the data or the schema without guessing. We speak Draft 2020-12, which is the default, plus Draft 2019-09 and Draft 7. Send both sides as strings, or as file references when they are large enough that escaping them into JSON would be painful. Accepts file input via file_ref — no need to extract file content first.

Deciding whether a document matches a contract is the check everything else rests on, and hand-rolling it means reimplementing a specification. You get a specification-compliant answer with the failing location named, every time.

When to use it

Scope

Non-goals

• We never work out a schema from your data — that is schema.infer_json_schema
• We never pull in another document, so send a self-contained schema
• We never merge, rewrite, or repair the schema or the document

Properties

Capabilities

Cost & billing

Flat rate per invocation

File handling

File limits

Constraints this skill applies to your input and generated files. The “Limited by” column shows the static source of each cap; none of these can be raised by upgrading your plan.

How we handle your data

Input schema

Output schema

API contract

curl -sS -X POST "$BASE/v1/execute" -H "Authorization: Bearer $KEY" -H 'Content-Type: application/json' -d '{"skillKey":"schema.validate_json","input":{"schema":"{\"type\":\"object\"}","instance":"{\"name\":\"Alice\"}"}}'

import { Squillify } from "@squillify/sdk";

const sq = new Squillify({ apiKey: process.env.SQUILLIFY_API_KEY });

const { output } = await sq.execute("schema.validate_json", {"instance":"{\"name\":\"Alice\",\"age\":30}","schema":"{\"type\":\"object\",\"properties\":{\"name\":{\"type\":\"string\"},\"age\":{\"type\":\"integer\"}},\"required\":[\"name\",\"age\"]}"});

Errors

Other limits

Limitations

• We never work out a schema from your data — that is schema.infer_json_schema
• We never pull in another document, so send a self-contained schema
• We never merge, rewrite, or repair the schema or the document
• Working out a schema from sample data (schema.infer_json_schema)
• Comparing two versions of a schema (schema.diff_json_schema)
• Schemas that depend on fetching another document

How to call

POST /v1/execute (requires authentication)

{
  "skillKey": "schema.validate_json",
  "input": {
    "instance": "{\"name\":\"Alice\",\"age\":30}",
    "schema": "{\"type\":\"object\",\"properties\":{\"name\":{\"type\":\"string\"},\"age\":{\"type\":\"integer\"}},\"required\":[\"name\",\"age\"]}"
  }
}

Examples

{
  "instance": "{\"name\":\"Alice\",\"age\":30}",
  "schema": "{\"type\":\"object\",\"properties\":{\"name\":{\"type\":\"string\"},\"age\":{\"type\":\"integer\"}},\"required\":[\"name\",\"age\"]}"
}

{
  "valid": true,
  "errorCount": 0,
  "draft": "https://json-schema.org/draft/2020-12/schema",
  "errors": []
}

{
  "instance": "{\"name\":\"Bob\"}",
  "schema": "{\"type\":\"object\",\"properties\":{\"name\":{\"type\":\"string\"},\"age\":{\"type\":\"integer\"}},\"required\":[\"name\",\"age\"]}"
}

{
  "valid": false,
  "errorCount": 1,
  "draft": "https://json-schema.org/draft/2020-12/schema",
  "errors": [
    {
      "path": "",
      "schemaPath": "/required",
      "message": "required property 'age' not found",
      "keyword": "required"
    }
  ]
}

Recommended use cases

• Check a request or response against the contract before you act on it
• Stop a bad document at the edge of an ingestion pipeline
• Assert in a test that two services still agree on a payload shape

Avoid for

• Working out a schema from sample data (schema.infer_json_schema)
• Comparing two versions of a schema (schema.diff_json_schema)
• Schemas that depend on fetching another document

Related skills

Composable with

Safety & privacy

• Runs without calling any external service.
• Your payloads are never stored permanently.
• Telemetry captures metadata only — never your payload contents.
• We never train models on your payloads.
• No human reviews your payloads.

Tags