We are attempting to build a Terraform data source that reads participant attributes from a live voice conversation via the Genesys Cloud Conversations API. The endpoint in question is GET /api/v2/conversations/{conversationId}/participants. The goal is to expose these attributes as a read-only variable in our state for downstream orchestration. The payload received consistently returns an empty attributes object, even though the attributes are clearly visible in the Genesys Cloud Admin UI and are being set by an external system via the PATCH /api/v2/conversations/{conversationId}/participants endpoint. The HTTP status code is 200 OK, but the JSON response for the specific participant looks like this: {“id”: “12345”, “attributes”: {}}. We have verified that the external system is correctly sending the attributes in the PATCH request body with the correct media type and authentication token.
The Terraform configuration uses a custom HTTP data source to fetch the participant details. The code snippet for the data source looks like this: data “http” “conversation_participant” { url = “https://${var.genesys_host}/api/v2/conversations/${var.conversation_id}/participants” request_headers = { Authorization = “Bearer ${var.genesys_token}” Accept = “application/json” } }. The issue seems to be that the GET endpoint does not return the custom attributes set by the external system, only the default attributes. We have also tried using the GET /api/v2/conversations/{conversationId} endpoint to get the full conversation details, but the participant attributes are still empty. The documentation states that participant attributes can be read and written via the API, but it does not specify any limitations on reading custom attributes set by external systems. We are unsure if this is a bug in the API or if there is a specific configuration required to enable reading of custom attributes.