POST
/
agents
/
{id}
/
query

​
Streaming

When streaming is enabled, the endpoint will emit events “answer” (answer of the model) and “endpoint_response” (full response of the endpoint)
import {
  EventStreamContentType,
  fetchEventSource,
} from '@microsoft/fetch-event-source';

let buffer = '';
let bufferEndpointResponse = '';
const ctrl = new AbortController();

await fetchEventSource(queryAgentURL, {
        method: 'POST',
        headers: {
          'Content-Type': 'application/json',
        },
        signal: ctrl.signal,
        body: JSON.stringify({
          streaming: true,
          query,
          conversationId,
          visitorId,
        }),

        async onopen(response) {
          if (response.status === 402) {
            throw new ApiError(ApiErrorType.USAGE_LIMIT);
          }
        },
        onmessage: (event) => {
          if (event.data === '[DONE]') {
            // End of stream
            ctrl.abort();

            try {
              const { sources, conversationId, visitorId } = JSON.parse(
                bufferEndpointResponse
              ) as ChatResponse;
            } catch {}
          } else if (event.data?.startsWith('[ERROR]')) {
            // ...
          } else if (event.event === "endpoint_response") {
            bufferEndpointResponse += event.data;
          } else if (event.event === "answer") {
            buffer += event.data;
            // ...
          }
       },
  });

Authorizations

​
Authorization
string
header
required

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

Path Parameters

​
id
string
required

ID of the agent

Body

application/json
​
query
string
required

This is the query you want to ask your agent.

​
conversationId
string

ID of the conversation (If not provided a new conversation is created)

​
filters
object
​
frequencyPenalty
number

Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.

​
maxTokens
number

The maximum number of tokens to generate in the chat completion.

​
modelName
enum<string>

Override agent model

Available options:
gpt_3_5_turbo,
gpt_3_5_turbo_16k,
gpt_4
​
presencePenalty
number

Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.

​
promptTemplate
string
deprecated

(DEPRECATED in favor of systemPrompt and userPrompt) Set the prompt template for this query

​
promptType
enum<string>
deprecated

(DEPRECATED in favor of systemPrompt and userPrompt) Set the prompt type for this query

Available options:
raw,
customer_support
​
streaming
boolean

Enable streaming

​
systemPrompt
string

Agent system prompt

​
temperature
number

Temperature of the model (min 0.0, max 1.0)

​
topP
number

An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or temperature but not both.

​
userPrompt
string

Agent user prompt

​
visitorId
string

ID of the participant that's sending the query (If not provided a new ID is created)

Response

200 - application/json
​
answer
string

The answer of the agent.

​
conversationId
string

ID of the conversation

​
sources
object[]
​
visitorId
string

ID of the participant that's sending the query

Was this page helpful?

Suggest editsRaise issue