Concept
Chats is the group of endpoints focused on real-time human support operations — what happens when a human agent takes over a conversation that was being managed by AI. While the Conversations group provides a complete view of history (reading, searching, analysis), the Chats group is action-oriented: sending messages on behalf of a human agent, starting and stopping human mode, and deleting messages from active chats. The practical difference: use Conversations to access and analyze historical data; use Chats when you are building a real-time human support interface — such as a custom inbox, a support panel, or an integration with a help desk system.Handoff: How It Works
When a human agent needs to take over a conversation from the AI:Main Chat Fields
| Field | Type | Description |
|---|---|---|
id | string (UUID) | Chat ID (same as conversation_id) |
status | enum | open | human | closed |
assigned_to | string (UUID) | ID of the active human agent |
human_started_at | string | ISO 8601 handoff start timestamp |
channel_id | string (UUID) | Channel through which messages are sent |
contact_id | string (UUID) | Contact being attended to |
Chat Message Fields
| Field | Type | Description |
|---|---|---|
id | string (UUID) | Message ID |
conversation_id | string (UUID) | Conversation it belongs to |
content | string | Text content |
type | enum | text | image | audio | video | document |
sender_type | enum | agent (AI) | human | contact |
sender_id | string (UUID) | ID of the human or agent sender |
status | enum | sent | delivered | read | failed |
created_at | string | ISO 8601 send timestamp |
Available Endpoints
See the left sidebar for the full list of endpoints in this group. Each endpoint has its own page with request/response schema, cURL/JS/Python examples, and an interactive playground.
Next Steps
Your first request
If you haven’t made your first call yet, start here.