Rynko
Rynko
Sign InGet Started
Rynko Flow

Deterministic Validation That Never Lies

Schema constraints and expression-based business rules. No AI, no uncertainty — just structured rules that pass or fail.

JSON Schema validation
Sub-35ms response
Expression engine
Start ValidatingView Docs
Schema Validation

Structural Guarantees for Every Field

Define what your data must look like. Types, constraints, required fields, nested objects, arrays with item schemas — all declarative.

string
Text values with minLength, maxLength, pattern, format, allowedValues
number
Numeric values with min, max, integer constraints
boolean
True/false flags
date
ISO 8601 dates with min/max range validation
array
Lists with itemType, minItems, maxItems constraints
object
Nested structures with their own property schemas
Example: Invoice Schema
{
  "invoice_number": {
    "type": "string",
    "required": true,
    "pattern": "^INV-[0-9]{6}$"
  },
  "total": {
    "type": "number",
    "required": true,
    "min": 0
  },
  "currency": {
    "type": "string",
    "allowedValues": ["USD", "EUR", "GBP"]
  },
  "line_items": {
    "type": "array",
    "required": true,
    "minItems": 1,
    "itemType": "object",
    "properties": {
      "description": { "type": "string", "required": true },
      "quantity": { "type": "number", "min": 1 },
      "unit_price": { "type": "number", "min": 0 }
    }
  }
}
Business Rules

Cross-Field Logic Without Code

Write expressions that reference any field in your payload. Arithmetic, comparisons, string methods, array operations, and custom error messages.

Operators

Arithmetic: +, -, *, /, %

Comparison: ==, !=, >, <, >=, <=

Logical: &&, ||, !

Built-in Functions

Aggregation: sum(), Math.round(), Math.max(), Math.min()

String: .toUpperCase(), .toLowerCase(), .trim(), .length, .includes(), .startsWith(), .endsWith()

Array: map, filter, find, some, every, reduce

Example Rules

Total matches line items

Cross-field validation: computed total must match the sum of line item amounts.

total == sum(line_items, 'quantity * unit_price')
Discount within bounds

Business constraint: discount cannot exceed 30%.

discount_percent >= 0 && discount_percent <= 30
Ship date after order date

Temporal logic: shipment cannot precede the order.

ship_date > order_date
Custom error message

The fail() function lets you return a specific error message when a rule trips.

quantity > 0 || fail('Quantity must be positive')
HS code from lookup table

Reference external data at validation time using lookup tables.

lookup('hs_codes', hs_code) != null || fail('Invalid HS code: ' + hs_code)
Lookup Tables

Reference Data at Validation Time

Upload CSV/JSON reference data and query it from your rules. HS codes, sanctions lists, port codes, currency tables — anything your validation logic needs.

HS Code Validation
Check that HS codes exist in the official tariff schedule.
Sanctions Screening
Screen entity names against sanctions and denied-party lists.
Port & Country Codes
Validate UN/LOCODE port codes and ISO country codes.
Custom Reference Data
Upload any key-value dataset. CSV or JSON. Sync via API.
Using lookup() in Rules
// Validate HS code exists in tariff schedule
lookup('hs_codes', hs_code) != null
  || fail('Unknown HS code: ' + hs_code)

// Check country is not sanctioned
lookup('sanctioned_countries', country_code) == null
  || fail('Sanctioned country: ' + country_code)

// Get tax rate from reference table
tax_amount == subtotal * lookup('tax_rates', region)
How It Works

Three Steps to Validated Data

Define your schema, add business rules, submit data. Deterministic results in under 35ms.

1
Define Your Schema
Specify field names, types, and constraints. Required fields, allowed values, string patterns, numeric ranges — all declarative.
2
Add Business Rules
Write cross-field expressions, lookup table checks, and custom fail messages. Pure logic, no code to deploy.
3
Submit Data
POST JSON to the gate endpoint. Get pass/fail with detailed error traces in under 35ms. No AI, no latency surprises.

The Validation Pipeline

Schema

Structural validation — types, constraints, required fields

Business Rules

Cross-field logic, lookups, custom error messages

AI Judge

Optional — semantic evaluation for what rules can't check

Ready to Validate?

Define your schema, add business rules, and start validating agent outputs in minutes. Deterministic. Fast. No AI required.

Start ValidatingRead the Docs