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: APIClarity
providerId: apiclarity
generated: '2026-09-04'
method: searched
source: >-
  https://raw.githubusercontent.com/openclarity/apiclarity/master/api3/global/openapi.gen.yaml,
  https://github.com/openclarity/apiclarity#readme
modified: '2026-09-04'
created: '2026-05-04'
description: >-
  APIClarity publishes no rate limits, and implements none. It is a single-tenant control plane
  you run in your own cluster — there is no vendor quota to exceed and no shared capacity to
  protect. No operation declares a 429, no response header carries a limit, and no documentation
  describes throttling.
limit_count: 0
limits: []
headers: {}
headers_note: >-
  No X-RateLimit-*, no RateLimit-*, no Retry-After. Verified against all 65 operations in the
  aggregated specification and against every module contract.
status_on_exhaustion: null
status_on_exhaustion_note: No 429 response is declared anywhere in any APIClarity contract.
evidence:
- url: https://raw.githubusercontent.com/openclarity/apiclarity/master/api3/global/openapi.gen.yaml
  status: 200
  finding: 65 operations; zero declare a 429 and zero declare a rate-limit response header.
- url: https://github.com/openclarity/apiclarity#readme
  status: 200
  finding: No throttling, quota or fair-use section.
related_capacity_controls:
  note: >-
    APIClarity does expose ingest-side controls, which are capacity settings on your own
    deployment rather than API rate limits. They are recorded here so the distinction is not
    lost — trace sampling is configurable, and release v0.14.4 fixed "trace enablement handling
    when sampling is disabled" and "limit application for annotation events". These govern how
    much traffic the deployment ingests, not how often a client may call it.
correction:
  date: '2026-09-04'
  note: >-
    This file previously declared X-RateLimit-Limit / X-RateLimit-Remaining / Retry-After
    headers, three invented per-tier limits (100 req/min free, 100,000 req/month professional,
    1,000 req/min enterprise with a 5,000 burst) and four invented policies including a fair-use
    clause and a UTC monthly quota reset. None of it existed; it came from a 2026-05-04 bulk
    sweep marked `method: generated`. Replaced with the measured zero.