Model Library

A browser-first tool for a versioned, reusable model / component library — the API design-governance primitive Spectral leaves out. Author or import a library of named JSON Schemas with semantic versions, see which OpenAPI specs consume each model, and when a model version bumps, know instantly whether the change is breaking or non-breaking for every consumer — plus a drift detector that surfaces near-duplicate model variants. No backend, no accounts; runs entirely in your browser. Live at library.apicommons.org.

Linters check one document at a time. They cannot tell you that forty APIs each define their own slightly-different User, or that bumping the shared Money model quietly breaks the three services that consume it. The missing capability is a place for the models themselves to live, version, and be reused — so specs $ref a canonical definition instead of copying it.

The library

A model library is a small, machine-readable file — a version plus a list of reusable named schemas, each with its own semantic version, owner, and description:

version: "1.0"
models:
  - name: User
    version: "2.0.0"
    owner: team-identity
    description: A platform user account.
    schema:
      type: object
      required: [id, email, name]
      properties:
        id:     { type: string, format: uuid }
        email:  { type: string, format: email }
        name:   { type: string }
        status: { type: string, enum: [active, disabled] }

Specs consume a model by $ref-ing #/components/schemas/<Name> (or, as a drift signal, by defining an inline schema of the same name).

What it reports

Paste your library and the OpenAPI spec(s) that reference it, and it produces three views:

• Models — every named schema and version in the library, with its owner, property count, and how many specs consume it. Two rows for one name are versions to reconcile.
• Change propagation — a structural diffSchema(old, new) of each version bump, with every change classified breaking (a removed/renamed required property, a narrowed type, a removed enum value) or non-breaking (an added optional property, an added enum value, a description edit) — and the verdict propagated to each spec that $refs the model.
• Drift & dedup — near-duplicate model variants (same-ish name, or property-set overlap by Jaccard ≥ 0.6) across the library and the inline schemas your specs define, so you can see "three variants of a user model" and collapse them onto one definition.

Develop

npm install
npm run dev
npm run build     # → dist/

Pure client-side; no data build. The samples in public/ demonstrate a breaking version bump, non-breaking changes, and a multi-variant drift cluster.

Privacy

Everything runs client-side. The library and specs you paste never leave the page — there is no server.

---

This tool came out of the State-of-Spectral research and a conversation with a Spectral maintainer about the capability API governance tooling still misses: not another linter, but a versioned model / design library where reusable schemas live, version, and are reused.

A project of API Evangelist, maintained openly under API Commons. Free to fork; API Evangelist offers expert API design and governance services — including standing up a real shared model library — when you want help. Apache-2.0.

Part of API Commons

An open, browser-first tool from API Commons — free, no backend, your data stays in your browser. Browse the full set at apicommons.org/tools.

Related tools

• Spec Review — ref-resolving breaking-change diff for PRs
• API Validator — lint OpenAPI/AsyncAPI/Arazzo/JSON Schema in-browser
• Code-First Governance — govern generated specs; fix findings in code
• API Governance Graph — bind building blocks into one graph + Gaps view
• Spectral Ruleset Studio — turn a style guide into an owned ruleset
• API Reusability — score API reuse across an org

License

Apache-2.0.

API Commons licenses code under Apache-2.0 and artifacts — schemas, rulesets, examples and API descriptions — under CC BY-NC-SA 4.0.