generated: '2026-08-12'
method: searched
source: >-
  https://docs.skyvia.com/api-reference/ + https://skyvia.com/pricing +
  https://docs.skyvia.com/account-management/subscription-limits-and-plans-in-more-details.html +
  openapi/_original/skyvia-public-api-openapi-original.json
limit_count: 0
summary: >-
  Skyvia publishes NO per-request rate limit for any of its APIs, and NO rate-limit response headers. The
  API reference has no rate-limit section; the 52-operation Swagger document declares no 429 response and no
  headers on any response; and no RateLimit-*, X-RateLimit-* or Retry-After header was observed on live
  responses. An honest zero: a client has no runtime signal telling it how close it is to a ceiling, and no
  documented status code to back off on.
response_headers: []
status_on_exhaustion: null
retry_after: false
limits: []
what_is_metered_instead:
  note: >-
    Skyvia meters subscription-level THROUGHPUT rather than request rate. These are real, published ceilings
    that an API or agent caller will hit, but they are billing quotas enforced per month or per day against
    the account, not per-request limits with a runtime signal. Captured in full in
    plans/skyvia-plans-pricing.yml.
  quotas:
  - product: Data Integration
    metric: records processed per month
    free: 10000
    paid: unlimited
    consumed_by: POST /v1/workspaces/{workspaceId}/integrations/{integrationId}/executions
  - product: Data Integration
    metric: minimum scheduling frequency
    free: once a day
    paid: once a day (Basic) / once hourly (Standard) / once per minute (Professional) / instant (Enterprise)
    note: >-
      This is the closest analogue to a rate limit on the platform — it caps how often a scheduled
      integration may run, by plan.
  - product: Automation
    metric: tasks per month
    free: 1000
    paid: unlimited
    consumed_by: webhook, schedule and connection triggers
  - product: Automation
    metric: minimum schedule interval
    free: 1 day
    paid: 1 minute
  - product: Query
    metric: queries per day
    free: 5
    paid: unlimited
  - product: Connect
    metric: traffic per month (bytes)
    free: 100 KB
    paid: 1 MB (Basic) / 100 MB (Standard) / 1 GB (Professional)
    consumed_by: OData endpoints, SQL endpoints, and MCP endpoint tool calls
    note: >-
      The binding constraint on the agent surface. Because it is measured in bytes rather than calls, the
      cost of an MCP Execute call scales with result-set size, and an unbounded query can exhaust a month's
      quota in one request.
  - product: Backup
    metric: storage
    free: 1 GB
    paid: 20 GB / 200 GB / 1 TB
connector_side_limits:
  note: >-
    Skyvia does throttle OUTBOUND requests to the third-party APIs it integrates with, to stay inside those
    vendors' limits — the August 2026 release note records rate-limit throttling added to the Cin7 Core
    Inventory connector "to regulate request frequency in compliance with the Cin7 Core Inventory API
    limits", and community threads discuss Salesforce Bulk API limits. That is Skyvia respecting other
    people's limits, not Skyvia publishing its own.
  evidence:
  - https://docs.skyvia.com/recent-releases/august-2026.html
mcp_surface_controls:
  note: >-
    The Connect MCP Execute tool carries per-call bounds in its own inputSchema that function as soft limits
    even though they are caller-set rather than server-enforced ceilings.
  controls:
  - name: pageSize
    default: 500
    description: Maximum records returned per page.
  - name: commandTimeout
    default: null
    description: Command timeout in seconds.
  - name: expireTimeout
    default: 3600
    description: Result-reader lifetime in seconds.
gaps:
- No documented request-rate limit on the Public REST API, Connect endpoints, or the MCP server.
- No 429 declared in the spec and no rate-limit or Retry-After headers observed.
- No published guidance on concurrency, burst behaviour, or backoff.
evidence:
- url: https://docs.skyvia.com/api-reference/
  status: 200
  finding: no rate-limit section; grep for "rate limit", "Retry-After", "X-RateLimit" returned nothing
- url: https://api.skyvia.com/swagger/v1/swagger.json
  status: 200
  finding: no 429 response and no response headers declared on any of 52 operations
- url: https://skyvia.com/pricing
  status: 200
  finding: quotas published per product per month; no per-request limits