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: Juniper Networks
providerId: juniper
generated: '2026-09-18'
method: searched
source: >-
  openapi/juniper-mist-api-openapi.yml (the 429 response description in Mist API 2607.1.1),
  https://www.juniper.net/documentation/us/en/software/mist/api/http/guides/api-requests/rate-limit,
  https://www.mist.com/documentation/api-rate-limiting/,
  plus a live unauthenticated probe of https://api.mist.com/api/v1/login on 2026-09-18.
created: '2026-05-04'
modified: '2026-09-18'
reconciled: true
supersedes_note: >-
  Replaces the 2026-05-04 bulk-sweep artifact, which recorded "authoritative numbers are
  gated behind authenticated docs". They are not gated — the number is written into the
  provider's own published OpenAPI, in the description of the 429 response on all 1,058
  operations that declare one.
description: >-
  Rate limits for the Juniper Mist API, the one Juniper surface with a published limit.
  5,000 calls per hour per API token, reset on the hourly boundary, 429 on exhaustion with
  a Retry-After header the contract does not document.
tags:
  - AI
  - Automation
  - Cloud
  - Enterprise
  - Networking
  - SDN
  - Security
  - Rate Limiting
sources:
  - https://www.juniper.net/documentation/us/en/software/mist/api/http/guides/api-requests/rate-limit
  - https://www.mist.com/documentation/api-rate-limiting/
  - openapi/juniper-mist-api-openapi.yml
responseCodes:
  throttled: 429
  serviceUnavailable: 503
limit_count: 1
limits:
  - name: Mist API calls per token
    api: Mist API
    scope: per-api-token
    metric: requests
    limit: 5000
    window: 1h
    burst: null
    reset: >-
      Hourly boundary — "The current rate limiting is 5000 API calls per hour and is reset
      at the hourly boundary."
    status_on_exhaustion: 429
    error_detail: >-
      "Too Many Request. The API Token used for the request reached the 5000 API Calls per
      hour threshold"
    error_source: >-
      Quoted verbatim from the 429 response description in the published OpenAPI — the
      strongest evidence available, since it is in the contract rather than in prose.
    increase: Contact Juniper/Mist support (support@mist.com) with the use case.
headers:
  documented: []
  documented_note: >-
    The contract declares an EMPTY headers object on every 429 response. Nothing about the
    runtime rate-limit signal is specified.
  observed:
    - name: Retry-After
      value_observed: '73'
      status: 429
      probed_url: https://api.mist.com/api/v1/login
      probed: '2026-09-18'
      note: >-
        Returned in practice but undocumented. An agent should honour it when present and
        fall back to waiting for the hourly boundary when it is not.
  absent:
    - X-RateLimit-Limit
    - X-RateLimit-Remaining
    - X-RateLimit-Reset
    - RateLimit
    - RateLimit-Policy
  absent_note: >-
    No RFC 9238-style RateLimit headers and no X-RateLimit-* family were observed on a
    live 429. A client cannot see how much budget is left before it runs out — only that
    it has run out.
event_rate_limits:
  source: https://www.juniper.net/documentation/us/en/software/mist/automation-integration/topics/concept/webhooks-and-alerts.html
  limits:
    - name: Rogue AP reporting
      limit: once per 10 hours
      scope: webhook event
    - name: Rogue client / Honeypot AP events
      limit: once per 10 minutes
      scope: webhook event
other_products:
  note: >-
    No public rate limits are published for Junos Space, Apstra, Paragon/Routing Director
    or the Junos device REST/NETCONF surfaces. Those are customer-hosted controllers whose
    throughput is a function of the operator's own deployment, which is a reason there is
    no number rather than an omission.
policies:
  - name: Backoff
    description: >-
      Exponential backoff with jitter on 429/503, honouring Retry-After when present.
      Because no remaining-budget header exists, clients should also track their own call
      count against the 5,000/hour ceiling.
maintainers:
  - FN: Kin Lane
    email: kin@apievangelist.com