generated: '2026-08-15'
method: searched
source: https://docs.medable.com/cortex-api/scripting/script-limits.md
docs:
- https://docs.medable.com/cortex-api/scripting/script-limits
- https://docs.medable.com/getting-started/cortex-user-guide/making-requests
- https://docs.medable.com/cortex-api/faults/fault-reference
limit_count: 12
description: >-
  Medable does not publish a conventional per-key/per-second HTTP rate limit for the
  Cortex API, and returns no RateLimit-*/X-RateLimit-* response headers. What it DOES
  publish is a set of hard platform quotas: per-execution limits on server-side Cortex
  scripts (jobs, routes, triggers), a daily job-run cap, a request size ceiling, and a
  documented throttling fault family that surfaces as HTTP 429. Throttling is also
  applied by org Policy (cortex.throttled.policy). Limits below are the published
  DEFAULTS; the docs describe them as defaults that a multi-tenant environment enforces
  so runaway scripts cannot monopolize shared resources.
response_headers:
  rate_limit_headers: []
  retry_after: false
  notes: >-
    No RateLimit-*, X-RateLimit-* or Retry-After header is documented on any Cortex
    response. The only documented cross-cutting response header is Medable-Server-Time
    (clock-drift correction for signature auth) and Medable-Csrf-Token. Agents cannot
    read remaining quota from a response; exhaustion is only observable as a fault.
exhaustion:
  status: 429
  envelope: object=fault
  faults:
  - errCode: cortex.throttled.policy
    code: kThrottled
    status: 429
    reason: Policy throttled request.
  - errCode: cortex.throttled.stalledCursor
    code: kThrottled
    status: 429
    reason: Stalled cursor was closed.
  - errCode: cortex.throttled.unspecified
    code: kThrottled
    status: 429
    reason: Request throttled.
  - errCode: cortex.tooBusy.sandbox
    code: kTooBusy
    status: 429
    reason: Sandbox too busy.
  - errCode: cortex.tooBusy.unspecified
    code: kTooBusy
    status: 429
    reason: Server too busy.
  ref: errors/medable-error-codes.yml
limits:
- name: request_size
  scope: per-request
  limit: 1024
  unit: kilobytes
  window: null
  description: Default maximum HTTP request size for a Cortex API call.
  source: https://docs.medable.com/getting-started/cortex-user-guide/making-requests
  exhaustion_fault: cortex.tooLarge.unspecified (413)
- name: job_timeout
  scope: per-script-execution
  limit: 10000
  unit: milliseconds
  window: null
  description: Maximum execution time for a Cortex Job.
  source: https://docs.medable.com/cortex-api/scripting/script-limits
- name: route_timeout
  scope: per-script-execution
  limit: 5000
  unit: milliseconds
  window: null
  description: Maximum execution time for a Cortex Route (custom API endpoint).
  source: https://docs.medable.com/cortex-api/scripting/script-limits
- name: trigger_timeout
  scope: per-script-execution
  limit: 5000
  unit: milliseconds
  window: null
  description: Maximum execution time for a Cortex Trigger.
  source: https://docs.medable.com/cortex-api/scripting/script-limits
- name: notifications_per_execution
  scope: per-script-execution
  limit: 100
  unit: notifications
  window: null
  description: Maximum number of email, push and/or SMS notifications per script execution.
  source: https://docs.medable.com/cortex-api/scripting/script-limits
- name: callouts_per_execution
  scope: per-script-execution
  limit: 5
  unit: requests
  window: null
  description: Maximum number of outbound http-module requests per script execution.
  source: https://docs.medable.com/cortex-api/scripting/script-limits
- name: callout_timeout
  scope: per-callout
  limit: 30000
  unit: milliseconds
  window: null
  description: Maximum time for an outbound http-module request.
  source: https://docs.medable.com/cortex-api/scripting/script-limits
- name: callout_request_size
  scope: per-callout
  limit: 524.288
  unit: kilobytes
  window: null
  description: Maximum size of an outbound http-module request body.
  source: https://docs.medable.com/cortex-api/scripting/script-limits
- name: callout_response_size
  scope: per-callout
  limit: 524.288
  unit: kilobytes
  window: null
  description: Maximum size of an outbound http-module response body.
  source: https://docs.medable.com/cortex-api/scripting/script-limits
- name: execution_depth
  scope: per-script-execution
  limit: 4
  unit: levels
  window: null
  description: Maximum levels of nested Cortex script executions.
  source: https://docs.medable.com/cortex-api/scripting/script-limits
- name: max_operations
  scope: per-script-execution
  limit: 1000000
  unit: operations
  window: null
  description: Maximum operations across Jobs, Triggers and Routes.
  source: https://docs.medable.com/cortex-api/scripting/script-limits
- name: job_runs_per_day
  scope: per-org
  limit: 24
  unit: runs
  window: 1d
  description: Maximum number of Cortex Job runs per day.
  source: https://docs.medable.com/cortex-api/scripting/script-limits
other_quotas:
- name: cortex_script_size
  limit: 50000
  unit: characters
  description: Maximum character length of a single Cortex script.
  source: https://docs.medable.com/cortex-api/scripting/script-limits
- name: notification_expiry
  limit: 30
  unit: days
  description: Unacknowledged notifications expire after 30 days.
  source: https://docs.medable.com/cortex-api/objects/notifications
gaps:
- No published per-second / per-minute HTTP request rate for the Cortex API.
- No rate-limit response headers, so remaining quota is not machine-readable.
- Policy-based throttling thresholds are org-configured and not published.