We are building a custom agent desktop widget using the .NET SDK and trying to programmatically start a cobrowse session for an active voice call. The documentation suggests using the POST /api/v2/conversations/voice/{conversationId}/participants endpoint with a specific action type. We have verified the conversation ID and participant ID are correct. The code looks like this:
var action = new ParticipantActionRequest
{
Action = "cobrowse",
To = new List<ParticipantAddress> { new ParticipantAddress { Address = "cobrowse://init" } }
};
client.ConversationsApi.PostConversationsVoiceParticipants(conversationId, participantId, action);
When we execute this, the API returns a 501 Not Implemented error immediately. The error body just says “Action not supported”. We are not seeing any documentation on a separate cobrowse initiation endpoint in the Conversations API. We have checked the Embeddable Client App SDK docs, but that only covers the UI integration, not the backend API call. We need to trigger this from our custom C# logic, not just a button click in the standard desktop. Is there a specific header or payload format we are missing? The voice call is active and the participant is in a talking state. We have tried changing the address to just “cobrowse” but get the same 501. We are stuck on this implementation detail. The SDK method is throwing the exception right at the API call level. We have confirmed our OAuth token has the necessary scopes for conversation control. The issue seems to be the action type itself. We are unsure if cobrowse is even exposed via this specific endpoint or if it requires a different path entirely. We have searched the API reference but found nothing on cobrowse initiation via REST.