Screen recording metadata missing in legal hold exports

What is the standard approach to retrieve the full chain of custody metadata for screen recordings when using the bulk export api?

we are currently using the /api/v2/recordings/search endpoint to identify recordings for a legal discovery request. the environment is genesys cloud v2024.2 in the europe/london region. when we trigger the export job via /api/v2/bulkexport/jobs, the resulting s3 bucket contains the screen recording files (.mp4) and the basic metadata json files. however, the metadata json files for screen recordings are missing critical fields required for our audit trail, specifically the agent_login_id and the interaction_id correlation.

voice recordings export correctly with all metadata, including wrap-up codes and skill group assignments. but the screen recordings only contain the recording_id, start_time, and end_time. this breaks our chain of custody validation process because we cannot programmatically link the screen activity back to the specific agent session without manual cross-referencing, which is not acceptable for legal hold.

we have verified that the recording settings in admin are set to include metadata in exports. we also checked the interaction metadata via the /api/v2/interactions/{id} endpoint, and the data exists there. it seems the bulk export job is not pulling the full interaction context for screen media types.

is there a specific parameter we need to add to the search query or the export job configuration to force the inclusion of agent and interaction metadata for screen recordings? or is this a known limitation with the current bulk export sdk version 2.1.0? we need a reliable way to ensure all exported media files have the necessary provenance data for compliance.

The root of the issue is that the /api/v2/recordings/search endpoint returns summary metadata, which intentionally excludes the granular chain-of-custody data required for strict legal compliance. That endpoint is designed for discovery, not forensic validation. To retrieve the full metadata, including the immutable audit trail and cryptographic hashes, you need to query the specific recording resource directly using /api/v2/recordings/{recordingId}.

When processing bulk exports, you must iterate through the search results and issue individual GET requests for each recording ID. The response body from this specific endpoint includes the metadata object containing the chainOfCustody array, which logs every access and export event with timestamps and actor IDs. This structure aligns with the ServiceNow REST API patterns I use for ticket auditing, where detailed context is fetched per record rather than in bulk.

Ensure your export job configuration includes the includeMetadata flag set to true, although the primary fix is switching the API call for data retrieval.

Note: The bulk export job itself does not modify the source recording metadata; it only copies files. Always validate the hash in the JSON against the MP4 file integrity.