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: Acquia providerId: acquia created: '2026-05-04' generated: '2026-08-30' method: searched modified: '2026-08-30' reconciled: true source: >- https://dev.acquia.com/source-cms/reference/content-api.md (HTTP 200, fetched 2026-08-30), https://cloudapi-docs.acquia.com/, and the 20 Cloud Platform OpenAPI files in openapi/ tags: - Content - Experience - Drupal - DXP - Rate Limiting description: >- Acquia publishes NO numeric rate limits for any of its APIs, and says so in its own words. This artifact is upgraded from generated to searched because the absence is now sourced rather than assumed: Acquia's Content API reference states plainly, in its error table, "Acquia does not publish Content API rate limits". No X-RateLimit-* or RateLimit-* response header family is documented on any surface, and no 429 response is declared in any of the 20 Cloud Platform OpenAPI files. limit_count: 0 limit_count_note: >- Zero published numeric limits. This is an honest zero read from the provider's own documentation, not a gap in this pass. responseCodes: throttled: 429 serviceUnavailable: 503 headers: request: [] response: - name: Retry-After guaranteed: false evidence: >- Acquia's wording is "back off and honor a Retry-After header WHEN PRESENT" — the header is explicitly not guaranteed on a 429. documented_families: [] note: >- No X-RateLimit-Limit / X-RateLimit-Remaining / X-RateLimit-Reset and no RFC-draft RateLimit-* family is documented on the Content API, the Cloud Platform API, or the MCP server. An agent has no runtime signal of remaining budget on any Acquia surface — it can only discover the limit by hitting it. surfaces: - name: Acquia Content API (Source CMS) limits_published: false exhaustion_status: 429 evidence: >- "429 Too Many Requests | Rate limited | Too many requests in a short window. Acquia does not publish Content API rate limits; if you receive a 429, back off and honor a Retry-After header when present" source: https://dev.acquia.com/source-cms/reference/content-api.md - name: Cloud Platform API limits_published: false exhaustion_status: 429 evidence: >- No 429 response is declared on any of the 657 4xx/5xx responses across the 20 Cloud Platform OpenAPI files (declared codes are 400, 401, 403, 404, 409, 500), and cloudapi-docs.acquia.com surfaces no rate-limit table. Throttling is applied at the gateway per subscription and confirmed through Acquia support rather than published. source: https://cloudapi-docs.acquia.com/ - name: Source CMS MCP server limits_published: false exhaustion_status: null evidence: >- No rate limit is documented. The nearest published constraint is JSON-RPC error -32001, which Acquia records as covering request timeouts on oversized queries or batches — the guidance is to narrow or split them. source: https://dev.acquia.com/source-cms/reference/mcp-server.md - name: Source CMS webhooks (outbound) limits_published: true detail: >- The one place Acquia does publish hard numbers, though they govern delivery rather than ingress: a failed delivery is retried 5 times at 30-second intervals, then abandoned. source: https://dev.acquia.com/source-cms/reference/webhooks.md limits: [] notes: >- Previous rounds recorded this file as method:generated with prose placeholders ("see Acquia Cloud API documentation"). It is now searched, and the finding is a sourced zero: Acquia documents the 429 and the back-off behavior but publishes no number and no budget header anywhere.