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: Red Hat Ansible Automation Platform
providerId: red-hat-ansible-automation-platform
generated: '2026-08-29'
method: searched
source: >-
  openapi/*.json in this repo, plus
  https://www.redhat.com/en/technologies/management/ansible/pricing and
  https://access.redhat.com/support/policy/updates/ansible-automation-platform
created: '2026-05-04'
modified: '2026-08-29'
supersedes: >-
  A 2026-05-04 bulk-sweep scaffold that asserted free/professional/enterprise tiers with
  10 rpm / 1,000 per month / 100 rpm / 100,000 per month / 1,000 rpm limits and a full
  X-RateLimit-* header set. None of that was published by Red Hat. It was fabricated scaffold
  data and has been replaced with the honest finding below.
description: >-
  Red Hat publishes NO API rate limits for Ansible Automation Platform. This is expected rather
  than a gap: AAP is software the customer installs and operates, so throughput is bounded by
  the customer's own control-plane sizing (instance groups, execution nodes, task capacity),
  not by a vendor quota. The one throttling signal anywhere in the contract is on Ansible
  Lightspeed, the hosted AI service.
limit_count: 0
limits: []
headers:
  documented: none
  detail: >-
    No RateLimit-*, X-RateLimit-* or Retry-After response header is declared in any of the
    1,767 operations across the five harvested OpenAPI documents.
responseCodes:
  throttled: 429
  throttled_scope: Ansible Lightspeed only
  evidence: >-
    14 operations in openapi/red-hat-ansible-automation-platform-ansible-lightspeed-openapi.json
    declare a 429 response described as "Request was throttled". No limit value, window or
    Retry-After header accompanies it.
capacity_model:
  kind: self-managed capacity, not vendor quota
  detail: >-
    AAP meters work through job execution capacity — instance groups, container groups,
    execution nodes and the `max_concurrent_jobs` / `max_forks` settings on instance groups —
    all of which are configured through the Controller API itself
    (controller.instance_groups_list / _create / _retrieve). An agent that needs to pace itself
    against AAP should read instance-group capacity, not a rate-limit header.
subscription_model:
  metric: managed nodes
  detail: >-
    Commercially AAP is sold per managed node, not per API call. Nothing in the subscription
    meters API requests.
  source: https://www.redhat.com/en/technologies/management/ansible/pricing
hosted_services_note: >-
  The Hybrid Cloud Console surfaces (console.redhat.com Automation Hub and Automation
  Analytics) sit behind Red Hat's shared console gateway. Anonymous probes returned
  {"errors":[{"status":401,"detail":"Missing Authentication","meta":{"response_by":"gateway"}}]}
  with no rate-limit headers; any console-side limits are undocumented publicly.
evidence:
- url: https://www.redhat.com/en/technologies/management/ansible/pricing
  status: 200
- url: https://console.redhat.com/api/automation-hub/v3/openapi.json
  status: 200
- url: https://console.redhat.com/api/automation-hub/v3/
  status: 401
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com
  url: https://apievangelist.com