Create chat agent

Create a text chat agent on a Retell LLM or conversation flow.

POSThttps://api.tried.com.au/create-chat-agent

Creates a chat agent for text conversations (chats, the website widget). A chat agent runs the same response engines as voice agents, so you can share one LLM or flow between a voice agent and a chat agent. Custom LLM WebSocket engines are not supported for chat.

Request#

Body#

Field	Type	Required	Description
`response_engine`	object	Yes	`{ "type": "retell-llm", "llm_id", "version"? }` or `{ "type": "conversation-flow", "conversation_flow_id", "version"? }`. The engine must exist. Without `version` the latest version is used on every chat.
`agent_name`	string	No	1 to 120 characters. Default `New chat agent`.
`post_chat_analysis_data`	object[]	No	Analysis fields extracted after each chat into `chat_analysis.custom_analysis_data`. Default `[]`.
`post_chat_analysis_model`	string	No	Model for post-chat analysis. Default `gpt-5.6-terra`.
`analysis_summary_prompt`	string	No	Custom prompt for `chat_summary`. Up to 2000 characters.
`analysis_successful_prompt`	string	No	Custom prompt for `chat_successful`. Up to 2000 characters.
`webhook_url`	string or null	No	URL for chat events. When `null`, the workspace webhook URL is used. Default `null`.
`webhook_events`	string[]	No	Any of `chat_started`, `chat_ended`, `chat_analyzed`. Default: all three. Each delivery's body is `{ "event", "chat" }` with the chat object.
`end_chat_after_silence_ms`	integer	No	End a chat after this long without activity, 60000 to 86400000. Default `3600000` (1 hour).
`auto_close_message`	string	No	Up to 500 characters. Added as the last agent message when a chat is closed for inactivity (`end_chat_after_silence_ms`).

Example request#

curl -X POST "https://api.tried.com.au/create-chat-agent" \
  -H "Authorization: Bearer $TRIED_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "agent_name": "Bright Dental chat",
  "response_engine": {
    "type": "retell-llm",
    "llm_id": "llm_9c8b7a6f5e4d3c2b1a0f9e8d"
  },
  "post_chat_analysis_data": [
    {
      "type": "boolean",
      "name": "booked",
      "description": "Did the user book an appointment?"
    }
  ],
  "end_chat_after_silence_ms": 1800000
}'

const res = await fetch("https://api.tried.com.au/create-chat-agent", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.TRIED_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "agent_name": "Bright Dental chat",
    "response_engine": {
      "type": "retell-llm",
      "llm_id": "llm_9c8b7a6f5e4d3c2b1a0f9e8d"
    },
    "post_chat_analysis_data": [
      {
        "type": "boolean",
        "name": "booked",
        "description": "Did the user book an appointment?"
      }
    ],
    "end_chat_after_silence_ms": 1800000
  }),
});
const data = await res.json();

import os, requests

res = requests.post(
    "https://api.tried.com.au/create-chat-agent",
    headers={"Authorization": f"Bearer {os.environ['TRIED_API_KEY']}"},
    json={
        "agent_name": "Bright Dental chat",
        "response_engine": {
            "type": "retell-llm",
            "llm_id": "llm_9c8b7a6f5e4d3c2b1a0f9e8d",
        },
        "post_chat_analysis_data": [
            {
                "type": "boolean",
                "name": "booked",
                "description": "Did the user book an appointment?",
            },
        ],
        "end_chat_after_silence_ms": 1800000,
    },
)
data = res.json()

Response#

201 Created

{
  "chat_agent_id": "chat_agent_6a7b8c9d0e1f2a3b4c5d6e7f",
  "agent_name": "Bright Dental chat",
  "response_engine": { "type": "retell-llm", "llm_id": "llm_9c8b7a6f5e4d3c2b1a0f9e8d" },
  "post_chat_analysis_data": [
    { "type": "boolean", "name": "booked", "description": "Did the user book an appointment?", "required": false }
  ],
  "post_chat_analysis_model": "gpt-5.6-terra",
  "webhook_url": null,
  "webhook_events": ["chat_started", "chat_ended", "chat_analyzed"],
  "end_chat_after_silence_ms": 1800000,
  "last_modification_timestamp": 1792310400000
}

Field	Type	Description
`chat_agent_id`	string	`chat_agent_` plus 24 hex characters.
`last_modification_timestamp`	integer	Epoch ms.

Chat agents have a single editable record. Published snapshots are separate: see Publish chat agent.

Errors#

Status	When
`400`	Validation failed, or the engine is `custom-llm` (`Chat agents support retell-llm and conversation-flow engines`).
`401`	Missing or invalid API key.
`403`	The key's IP allowlist or scopes reject the request. This route needs the `calls` scope.
`404`	The LLM or flow (or its `version`) does not exist.
`429`	Rate limit exceeded. See Rate limits.

See Errors for the body format.