generated: '2026-09-04'
method: searched
source: >-
  https://betterstack.com/docs/uptime/api, https://betterstack.com/tag/changelog,
  https://status.betterstack.com/, https://betterstack.com/terms, plus live probes of
  /api/v2 and /api/v3 on uptime.betterstack.com (2026-09-04).
provider: Better Stack
providerId: betterstack
versioning:
  scheme: uri-path
  current_versions:
    uptime: v2
    uptime_incidents: v3
    telemetry: v1
    errors: v1
  policy_published: false
  note: >-
    Better Stack publishes no written versioning policy. Two Uptime versions answer
    simultaneously — GET /api/v2/incidents and GET /api/v3/incidents both return 401 rather
    than 404 — so v2 is still routed, but nothing states which is current, which is
    superseded, or how long v2 will remain.
deprecation:
  policy_published: false
  policy_url: null
  sunset_header: false
  deprecation_header: false
  rfc8594: false
  note: >-
    No deprecation policy, no notice period, and no RFC 8594 Sunset or Deprecation header was
    observed on any probed response. https://betterstack.com/docs/uptime/api/deprecations/ and
    /docs/uptime/api/errors/ both 404. A consumer has no published warning mechanism for a
    breaking change.
  deprecated_operations: []
  deprecated_operations_note: >-
    No operation in this repo's four OpenAPI documents carries a `deprecated: true` flag.
sla:
  published: false
  url: https://betterstack.com/terms
  note: >-
    Better Stack sells uptime monitoring and SLA reporting as a product, and its pricing page
    lists an Enterprise tier, but no public API SLA or uptime commitment document was found on
    the marketing or docs surface. Contractual terms are at /terms.
status_page:
  url: https://status.betterstack.com/
  status: 200
  machine_readable: false
  note: >-
    Better Stack runs its own status page on its own product. The status page host is an SPA
    that answers 200 with the same HTML shell for every path, so no machine-readable status
    feed (/api/v2/status.json, /history.json) could be confirmed — an agent must scrape it.
  incident_history_captured: blogs/
  incident_history_note: >-
    This repo holds 21 status-page incident write-ups harvested into blogs/, dated 2025-04-20
    through 2026-08-09 — degraded telemetry queries, uptime dashboard degradation, DDoS on
    status pages, an AWS Direct Connect outage, and an rsyslog ingestion issue.
changelog:
  url: https://betterstack.com/tag/changelog
  status: 200
  scheme: numbered issues ("Changelog #N") with a publication date
  cadence: irregular
  see: changelog/betterstack-changelog.yml
lifecycle_signals:
  api_surface_growing: true
  note: >-
    The direction of travel is toward the agent surface, not the REST surface. Changelog #12
    (2025-09-15) shipped the MCP server; the MCP catalog now exposes ~106 tools across Uptime,
    Telemetry and Errors while the REST management APIs remain undocumented for rate limits,
    errors, deprecation and versioning, and have no SDK in any language.