specification: API Commons Rate Limits
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/RateLimits
provider: Schema.org
providerId: schema-org
created: '2026-05-04'
modified: '2026-08-13'
generated: '2026-08-13'
method: probed
source: >-
  Live unauthenticated probes of https://schema.org/ and
  https://validator.schema.org/validate on 2026-08-13, plus the maintainer statement in
  https://github.com/schemaorg/schemaorg/discussions/3261
tags:
  - Schema.org
  - Structured Data
  - Rate Limiting
  - Throttling
description: >-
  Rate-limit posture for Schema.org. This file replaces a 2026-05-04 bulk-sweep scaffold
  that asserted invented free/pro tiers, X-RateLimit-* headers and a 1,000-requests-per-month
  quota. None of that exists. Schema.org publishes no rate limits and returns no rate-limit
  headers on its vocabulary and documentation hosts. The one throttled surface is
  validator.schema.org, which is Google-operated, undocumented, and which the project asks
  people not to call programmatically.
limit_count: 0
headers:
  limit: null
  remaining: null
  reset: null
  retryAfter: null
  policy: null
observed_response_headers:
  - host: schema.org
    request: GET https://schema.org/
    status: 200
    rate_limit_headers_present: false
    headers_observed:
      - 'cache-control: public, max-age=600'
      - 'etag: "AxXBLg"'
      - 'expires: <now+600s>'
      - 'access-control-allow-origin: *'
      - 'server: Google Frontend'
    note: >-
      No X-RateLimit-*, no RateLimit-*, no Retry-After. Caching, not throttling, is the
      documented load-management mechanism for the static vocabulary files.
limits: []
surfaces:
  - surface: schema.org vocabulary + documentation files
    published_limits: false
    detail: >-
      No limits documented anywhere on schema.org, in the GitHub repository, or in
      /docs/developers.html. The vocabulary is distributed as static files under
      /version/<release>/ and the project encourages downloading the bulk files rather than
      fetching term pages repeatedly. robots.txt (probed 200) disallows only
      /docs/search_results.html.
  - surface: validator.schema.org
    published_limits: false
    throttled: true
    status_code_on_exhaustion: 429
    detail: >-
      Undocumented throttling. A Schema.org maintainer states in GitHub discussion #3261
      that validator.schema.org "is a service run and hosted by Google as a contribution to
      the project", that it has "a lot of safeguards against scraping", and that there are
      "no resources or plans for Google to make an API" — in response to a user reporting
      429 errors from repeated calls. No limit value, window, burst or Retry-After contract
      is published, so an integrator has no way to stay inside the limit by design.
    provider_guidance: >-
      Do not call it programmatically. The project points programmatic validators at
      https://github.com/google/schemarama instead.
retry_guidance: >-
  For the vocabulary files: use conditional GET with If-None-Match against the published
  ETag and honour the ten-minute cache-control window rather than polling. For validation:
  run schemarama locally instead of calling validator.schema.org.
note: >-
  limit_count: 0 is an honest measurement, not a gap. A vocabulary served as cacheable
  static files does not need per-consumer quotas, and Schema.org has never published any.