The storage endpoints let you list, retrieve, and delete files that have been saved to disk by the curl service. Saved files are organized under three symlinked directory structures — by-job, by-domain, and by-date — so the same physical file can be referenced through any of the three paths. Use these endpoints to audit downloads, look up files by domain or date, or clean up old content.
Retrieve a paginated list of every file currently saved to storage. Each entry includes its relative path, byte size, creation timestamp, and the originating job UUID when one can be parsed from the path. Files appear under their by-job, by-domain, and by-date paths, but all three resolve to the same physical file.
Stream the raw contents of a previously saved file. The response uses Content-Type: application/octet-stream, so it works for any file type. The path parameter accepts any of the three symlinked forms (by-job/..., by-domain/..., or by-date/...).
Permanently delete a file from storage. This action cannot be undone. Deleting a file removes only the underlying physical file — any other symlinked paths pointing to it will subsequently return 404.