YounndAI Domains

YounndAI Domains

The Schema Registry

Typed schemas for
every industry

Structured tags with typed fields, validation rules, and trust signals. Open standard. Managed registry.

Explore RegistryLearn the Format
31 Official Domains//Open Standard//Schema as a Service
YounndAI Domains
DomainsSchemaPricingDocs
  1. Docs
  2. Concepts
  3. Conformance Scoring
concepts

Conformance Scoring

How domain quality scores work — criteria, formula, badge levels, and what drives your score.

What Is the Score?

Every domain in the registry carries a conformance score between 0.0 and 1.0. This score measures how well a schema adheres to quality standards — it's a machine-readable confidence signal for consumers and AI agents.

ℹ️Note

The score is earned, not assigned. It's computed automatically by the registry's conformance engine every time a schema is published or updated.

What Drives the Score

The scoring engine evaluates two dimensions, weighted differently:

Record Coverage
40% of total score
Field Coverage
60% of total score

Within each dimension, points are awarded for:

Completeness

  • Every field has a type declared
  • Every tag has at least one field
  • No orphaned references or broken links between tags

Naming Quality

  • Short, clear field names (avoid customerFirstNameValue → prefer first_name)
  • Consistent kebab-case or snake_case throughout
  • Naming convention compliance with the YON specification

Documentation

  • Descriptions on every tag and every field
  • Context added to metadata (tag_context entries)
  • At least 1 use-case documented per domain
  • Relevant related domains linked (e.g., yon.health links to yon.insurance)
  • External standard links where applicable (ISO, FHIR, etc.)

Technical Validity

  • Parser roundtrip — the schema is parsed by @younndai/yon-parser and the output matches the input. If parsing fails, the schema gets a significant score penalty.
  • Field constraints are valid and non-contradictory (e.g., min ≤ max)
  • Enum values are properly formatted and non-empty

Badge Levels

Scores map to five badge levels:

Improving Your Score

Scores are recalculated on every publish. To improve:

  1. Add descriptions to all fields and tags
  2. Add tag_context entries to metadata
  3. Document at least one use-case
  4. Link to related domains
  5. Fix any parser roundtrip failures
  6. Ensure naming conventions are consistent

💡Tip

The fastest way to improve your score is to add descriptions and context. These two factors account for the majority of the field coverage dimension (60% of total).

Next Steps

Verified Badge
How the Verified badge works — identity + score attestation.
Trust Model
The three trust fields: tier, verified, and score.
← Lifecycle & PermanenceVerified Badge →

On This Page

What Is the Score?
What Drives the Score
Completeness
Naming Quality
Documentation
Technical Validity
Badge Levels
Improving Your Score
Next Steps
Registry
Schema
Pricing
API
GitHub
YounndAI
YounndAIYou and AI, unifiedBuilt withNollamaNollama
© — YounndAI™. You and AI, unified. (pronounced “yoon-dye”). All rights reserved.