"error": "ManifestParsingError",
"message": "Failed to parse screen recording metadata for conversation_id: conv_99887766. The 'recording_type' field is missing or invalid."
We are encountering a persistent issue with our bulk export jobs for legal discovery requests involving digital channels, specifically screen recordings initiated via the Webchat co-browsing feature. The environment is Europe/London, and we are using the standard BYOC configuration.
When initiating a bulk export via POST /api/v2/analytics/conversations/export with a filter for recording_type: screen, the job completes successfully, but the resulting manifest.json file contains truncated entries for specific conversations. These conversations were recorded using the latest version of the Genesys Cloud desktop agent (v10.5.2) on Windows 11 endpoints.
The error indicates a ManifestParsingError when our internal ingestion pipeline attempts to validate the chain of custody metadata. Upon inspecting the raw S3 bucket contents, the .mp4 files exist and are playable, but the associated metadata JSON objects lack the recording_type and media_source fields. This breaks our automated legal hold tagging process, which relies on these fields to distinguish between audio-only, video, and screen recording assets for regulatory compliance.
We have verified that the screen recording permissions are correctly configured in the Admin portal under Security > End User Data. The issue does not appear with standard voice recordings or video calls, only with the co-browsing screen capture feature. Is this a known limitation with the current API version, or is there a specific parameter required in the export request body to force the inclusion of screen recording metadata? We need to ensure full audit trail integrity for these exports.