/v1/workspaces/{workspace_slug}/audit-logs/exportAuthStream the workspace audit log as NDJSON or CSV, applying the same `?action=` / `?actor_id=` / `?since=` / `?until=` filters as the list. Newest first. Reading the export is itself audited.
The response is streamed: rows are walked in keyset batches and written as they are read, so neither the service nor the client holds the whole log. This is the SOC 2 / GDPR export the platform docs promise. Hard cap of 1,000,000 rows per call; a wider request is refused with 413 before any body is written, so a truncated file can never masquerade as a complete one.
Path parameters
workspace_slug*stringSlug of the workspace. Platform routes name the workspace by slug, positionally in the SDK (`platform.workspaces.retrieve("acme")`); the product APIs bind `workspace_id` instead.
Query parameters
formatenum (2)Defaults to `ndjson`.
actionstringExact-match filter on the action identifier, same rules as the list. Mail, Waitlist, Social and Auth actions are reported under the product name (`mail.domain.verified`); the older internal spellings (`envoi.*`, `rally.*`, `agora.*`, `heimdall.*`) are still accepted here and match the same rows.
actor_idstring · uuidExact-match filter on the acting account UUID, same rules as the list.
sincestring · date-timeInclusive lower bound on `created_at`, same formats as the list.
untilstring · date-timeInclusive upper bound on `created_at`, same formats as the list.
Response · 200
Streamed export. `application/x-ndjson` (one JSON object per line) or `text/csv` (RFC 4180, header row first).
Example
Request
GET https://api.platform-auth.productcraft.co/v1/workspaces/{workspace_slug}/audit-logs/export
Authorization: Bearer YOUR_TOKEN