Ask LeadGPT for agent action recommendations

Returns non-mutating LeadGPT action cards for the authenticated workspace. The route derives workspace access from apiKeyAuth and only uses the client workspace id for auth scoping.

POST

https://api.channelautomation.com/api/customer-hub/leadgpt/ask

Authorization

BearerAuth
optional
HTTP Bearer (JWT)

ApiKeyAuth
optional
API key in header

Request body JSON required

workspace_id
string
required
Workspace id used by apiKeyAuth to validate membership. Business logic uses req.user.workspace_id after authentication.
prompt
optional
string
Natural language LeadGPT prompt. Empty prompts default to "What should I do next?"
selectedContactId
optional
string
Optional contact id to summarize or draft against.

format: uuid

Returns

200

LeadGPT answer and ranked non-mutating action cards

400

Missing workspace access or invalid request

401

Authentication required

JSON

error
optional
string
success
optional
boolean

404

Selected contact is not available in the authenticated workspace

500

Internal server error

JSON

error
optional
string
details
optional
string
success
optional
boolean

Ask LeadGPT for agent action recommendations

curl --request POST \
  --url https://api.channelautomation.com/api/customer-hub/leadgpt/ask \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "workspace_id": "string",
  "prompt": "string",
  "selectedContactId": "1b5e8964-45aa-454a-bfa3-92a6f9bcf78f"
}
'

const url = 'https://api.channelautomation.com/api/customer-hub/leadgpt/ask';
const options = {
  method: 'POST',
  headers: {'Content-Type': 'application/json', Authorization: 'Bearer <token>'},
  body: JSON.stringify({
    workspace_id: 'string',
    prompt: 'string',
    selectedContactId: '1b5e8964-45aa-454a-bfa3-92a6f9bcf78f'
  })
};

fetch(url, options)
  .then(res => res.json())
  .then(json => console.log(json))
  .catch(err => console.error(err));

import requests

url = "https://api.channelautomation.com/api/customer-hub/leadgpt/ask"

payload = {
    "workspace_id": "string",
    "prompt": "string",
    "selectedContactId": "1b5e8964-45aa-454a-bfa3-92a6f9bcf78f"
}
headers = {
    "Content-Type": "application/json",
    "Authorization": "Bearer <token>"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

Response example

{
  "error": "Authentication required",
  "success": false
}

{
  "error": "Internal server error",
  "details": "string",
  "success": false
}