aichat_create_conversation_v2
Manage AI conversations: chat, retrieve, update, delete, and batch operations with model selection, state control, pagination, and async callbacks via the AiChat v2 endpoint.
Instructions
Create/manage conversations via AiChat v2 endpoint.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| id | No | Conversation ID. Required for retrieve/update/delete actions. | |
| async | No | Whether to process the request asynchronously. | |
| limit | No | Pagination limit for retrieve_batch. | |
| model | No | Model to use for the request. | gpt-4.1 |
| title | No | Conversation title. | |
| action | No | Operation to perform: chat (default), retrieve, retrieve_batch, update, or delete. | chat |
| offset | No | Pagination offset for retrieve_batch. | |
| preset | No | Optional preset model configuration. | |
| message | No | Message payload as a string or structured content array. | |
| user_id | No | Filter or associate by user ID. | |
| messages | No | Conversation messages for update/action workflows. | |
| question | No | Question text for chat action. | |
| stateful | No | Whether to use stateful conversation mode. | |
| max_turns | No | Maximum number of turns for conversation history. | |
| references | No | Optional list of reference sources. | |
| model_group | No | Provider group filter for retrieve_batch. | |
| callback_url | No | Webhook URL to receive async completion callbacks. | |
| tool_results | No | Tool call results for follow-up turns. | |
| allowed_skills | No | Optional list of skills the agent may use. | |
| application_id | No | Filter or associate by application ID. | |
| unattended_policy | No | Optional unattended-execution policy configuration. | |
| allowed_mcp_servers | No | Optional list of MCP servers the agent may use. |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |