Chat

/routes/chat/completion

Published @ 08/07/2026 by

author Fatos
Fatos, Founder @ answerly.io
POST https://api.answerly.io/routes/chat/completion

Generates a response to a chat conversation. This endpoint is OpenAI spec safe, allowing for seamless integration with existing OpenAI-compatible libraries and tools.

Note: Models and any other AI configurations cannot be passed in the API request body. All AI configuration must be done directly to the agent from the Answerly dashboard.

Request Body

APIKey string Required
Your API key. You can read more about how to get an API Key by clicking here.

agentId string Required
The ID of your specific agent.

You can find the Agent ID in your Agent Settings -> Settings tab, located all the way at the bottom of the page.

input array Required
A list of messages comprising the conversation so far, formatted to standard OpenAI specifications.

input[].role string Required
The role of the messages author. Typically user, assistant, or system.

input[].content string Required
The contents of the message.

curl --request POST \
    --url 'https://api.answerly.io/routes/chat/completion' \
    --header 'Content-Type: application/json' \
    --data '{
        "APIKey": "YOUR_API_KEY",
        "agentId": "YOUR_AGENT_ID",
        "input": [
            {
                "role": "user",
                "content": "What is your name?"
            }
        ]
    }'

Return (success)

choices array
A list of chat completion choices containing the generated assistant role and content.

{
    "choices": [
        {
            "role": "assistant",
            "content": "I am an AI assistant created by Answerly."
        }
    ]
}

Return (failure)

error object
The error object containing failure details.

error.code string
An error code returned by the server. Possible values:
error
completions-error
unknown-error

error.message string
A human-readable message providing more details about the error.

{
    "error": {
        "code": "completions-error",
        "message": "The specified agentId could not be located."
    }
}