Quick Start - Sample IT Assistant

Let's learn how to use the main endpoints to interact with your sample Assistant.

Important: You must already have an account on our Alan platform.

Postman Collection

Don't forget! We have Postman collection that cover's our endpoints. You can find it on our side bar.

Get your API keys

Your API requests are authenticated using API keys. Any request that doesn't include an API key will get 403.

You can get your API key in Settings -> Development -> Development Token

Get your token

Make your first request

To make your first request, send an authenticated request to the pets endpoint. This will create a pet, which is nice.

Create Assistant

POST https://backend.az.misterturing.com/v4.0/assistant/

Request Body

Name	Type	Description
name*	string	Assistant name
description	string	The description of the main goal of the Assistant.

Request Body

{
    "name": "First Assistant",
    "description": "You're an Assistant that will generate random jokes in any language."
}

---

Response Body

201 created

{
    "external_id": "3fb69549-401d-442c-a76f-f2eec96a2b52",
    "name": "First Assistant",
    "description": "You're an Assistant that will generate random jokes in any language.",
}

400 Bad request

We return all information related to the Assistant, including the assistant_preferences , that determines how your iFrame youu be rendered. You can customize it via platform.

Take a look at how you might call this method via curl:

curl

curl --location 'https://backend.az.misterturing.com/v4.0/assistant/' \
--header 'authorization: Bearer your_development_token' \
--header 'content-type: application/json' \
--data '{
    "name": "First Assistant",
    "description": "You'\''re an Assistant that will generate random jokes in any language."
}'

Interacting with Assistant

To initiate a conversation with the Assistant you should first generate a preview version of it.

That means the Assistant will be configured using your latest changes on the Flows, Extractions and Folders.

To achieve that you can use the deploy endpoint. POST https://backend.az.misterturing.com/v4.0/assistant/deploy/

Request Body

Name	Type	Description
assistant_id*	string	Assistant external_id created above.

Request Body

{
    "assistant_id": "your_assistant_external_id"
}

---

Take a look at how you might call this method via curl:

curl

curl --location 'https://backend.az.misterturing.com/v4.0/assistant/deploy/' \
--header 'Authorization: Bearer your_development_token' \
--header 'Content-Type: application/json' \
--data '{
    "assistant_id": "your_assistant_external_id"
}'

Now we can go to the next step: create the Conversation.

Conversation

Conversation is how the Assistant maintains the context of the messages exchanged between the user and the assistant.

POST https://backend.az.misterturing.com/v4.0/assistant/conversation/

Request Body

Name	Type	Description
assistant_id*	string	Assistant external_id created above.

Request Body

{
    "assistant_id": "your_assistant_external_id"
}

Response

{
    "conversation_id": "UUID"
}

Second Tab

---

Take a look at how you might call this method via curl:

curl

curl --location 'https://backend.az.misterturing.com/v4.0/assistant/conversation/' \
--header 'authorization: Bearer your_development_token' \
--header 'Content-Type: application/json' \
--data '{
    "assistant_id": "d11ad9b2-5cfb-4d2d-a6c0-a100a0c9422c"
}'

Send a Message

Now with your Conversation created you send a message using the Chat endpoint.

POST https://backend.az.misterturing.com/v4.0/assistant/chat/

Request Body

Name	Type	Description
message*	string	Your message.
conversation_id	string	Previously created conversation_id.

Request Body

{
    "message": "Hey! Tell me a fun joke about AI!",
    "conversation_id": "UUID"
}

Chat Response

We stream back to you the response text of the Assistant response using the Content-type: text/event-stream; charset=utf-8

---

Take a look at how you might call this method via curl:

curl

curl --location 'https://backend.az.misterturing.com/v4.0/assistant/chat/' \
--header 'authorization: Bearer your_development_token' \
--header 'Content-Type: application/json' \
--data '{
    "message": "oi",
    "conversation_id": "bfe1d680-3488-43e5-9fa9-6b470382a7f5"
}'