POST
/
orchestrate
/
runs
/
stream
Chat With Orchestrate Agent As Stream
curl --request POST \
  --url http://localhost/api/v1/orchestrate/runs/stream \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "message": {
    "role": "<string>",
    "content": [
      {
        "response_type": "conversational_search"
      }
    ],
    "mentions": [
      {
        "type": "<string>",
        "id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
        "name": "<string>"
      }
    ],
    "document_ids": [
      "3c90c3cc-0d44-4b50-8888-8dd25736052a"
    ],
    "parent_message_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
    "additional_properties": {
      "wxa_message": {},
      "display_properties": {
        "skip_render": false,
        "is_async": false
      },
      "tool_calls": [
        {}
      ],
      "tool_call_id": "<string>",
      "tool_name": "<string>",
      "wxo_connection_status": {
        "connection_status": "<string>",
        "connection_message": "<string>"
      }
    },
    "assistant_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
    "context": {
      "values": []
    },
    "step_history": [
      {}
    ],
    "message_state": {}
  },
  "thread_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
  "llm_params": {
    "beam_width": 1,
    "decoding_method": "greedy",
    "include_stop_sequence": true,
    "length_penalty": {
      "decay_factor": 2,
      "start_index": 2
    },
    "max_new_tokens": 1,
    "min_new_tokens": 1,
    "random_seed": 2147483648,
    "repetition_penalty": 1.5,
    "return_options": {
      "generated_tokens": true,
      "input_parameters": true,
      "input_text": true,
      "input_tokens": true,
      "token_logprobs": true,
      "token_ranks": true,
      "top_n_tokens": 2
    },
    "stop_sequences": "[\" and \"]",
    "temperature": 1,
    "time_limit": 123,
    "top_k": 50,
    "top_p": 0.5,
    "truncate_input_tokens": 1,
    "typical_p": 0.505
  },
  "guardrails": {
    "hap": {
      "input": {
        "enabled": true,
        "threshold": 123
      },
      "output": {
        "enabled": true,
        "threshold": 123
      },
      "mask": {
        "remove_entity_value": true
      }
    },
    "social_bias": {
      "input": {
        "enabled": true,
        "threshold": 123
      },
      "output": {
        "enabled": true,
        "threshold": 123
      },
      "mask": {
        "remove_entity_value": true
      }
    },
    "pii": {
      "input": {
        "enabled": true,
        "threshold": 123
      },
      "output": {
        "enabled": true,
        "threshold": 123
      },
      "mask": {
        "remove_entity_value": true
      }
    }
  },
  "context": {
    "userID": 123,
    "userName": "User"
  },
  "agent_id": "<string>",
  "additional_parameters": {},
  "environment_id": "<string>",
  "version": 123,
  "context_variables": {}
}'
{
  "id": "<string>",
  "event": "run.started",
  "data": {}
}

Authorizations

​
Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Query Parameters

​
stream_timeout
integer
default:60000
​
multiple_content
boolean
default:false

Body

application/json
​
agent_id
string
required
​
message
object | null
​
thread_id
string | null
​
llm_params
object | null
​
guardrails
object | null
​
context
object | null

A key value object for each context variable you want to configure.

Example:
{ "userID": 123, "userName": "User" }
​
additional_parameters
object | null
​
environment_id
string | null
​
version
integer | null
​
context_variables
object | null

Context variables for the agent execution (e.g., {'wxo_email_id': 'user@domain.com', 'wxo_user_name': 'John Doe'})

Response

200
application/json

Successful Response

​
id
string
required
​
event
enum<string>
required
Available options:
run.started,
run.completed,
run.failed,
run.expired,
run.cancelled,
run.step.started,
run.step.completed,
run.step.failed,
run.step.delta,
message.created,
message.started,
message.delta,
message.completed,
run.step.thinking,
message.interrupt,
error,
done,
planning
​
data
object
required