wit-core

wit-core is a message processor for wit.ai.

Handle your intents by defining actions and responses.

Getting Started

Prerequisites

• Python 3.6+

Install wheel package.

pip install wheel

Instalation

pip install wit-core

Documentation

Domain

The domain.yaml defines how the chatbot should operate according to the definition of the intents.

In the domain you specify the intents, actions and responses.

greet:
    response: 
        text: "Hello, there!"

temperature_set:
    action: action_temperature
    response:
        text: "The temperature was changed to {action_return}"

order_pizza:
    action: action_pizza
    response:
        template: template_pizza

Actions

Actions are ways for you to write your own codes. For example, manipulating entities, make an API call, or to query a database.

def custom_action(resource):
    temperature = resource.get_entity("wit$temperature")

    return temperature.value

The resource parameter allows accessing the properties of the wit.ai response.

Resource properties:

get_latest_message

Returns the message sent by the user.

get_intent

Returns the properties of the intent.

get_entity("entity_name")

Returns the properties of the specified entity.

get_trait("trait_name")

Returns to the properties of the specified trait.

Responses

It is possible to define two types of responses: plain text and templates.

Text

You can directly in the domain specify in the response a text response quickly.

Templates

Templates serve to give you the possibility to add some logic to the answer. Templates receive the return value of an action.

def custom_template(action_return):
    # Template logic...

    return "Some response"

Command Line Interface

wit-core init

Creates the directory structure.

wit-core shell

Loads the domain and allows interaction with the chatbot.

wit-core http-server

Creates a http server that can be used for custom integrations. They provide a URL where you can post messages.

wit-core websocket-server

Create a websocket server for real-time interaction with the chatbot. They provide an endpoint where you can send messages.

How to use

Creates the folder structure needed for the project.

wit-core init

Create a .env at the root of the application with your secret wit.ai key.

# Wit.ai
ACCESS_TOKEN=YOUR-SECRET-KEY

Interact with the chatbot.

wit-core shell

Contributing

1. Fork it (https://github.com/yourname/yourproject/fork)
2. Create your feature branch (git checkout -b feature/fooBar)
3. Commit your changes (git commit -am 'Add some fooBar')
4. Push to the branch (git push origin feature/fooBar)
5. Create a new Pull Request

Development

Clone the repository:

git clone https://github.com/LucasOliveiraS/wit-core

Install the dependencies:

poetry install

Run tests:

make tests