Skip to main content

Build serverless chatbot on BotHub.Studio

Project description

This package provide components to works with BotHub.Studio, which is a chatbot hosting service.

With bothub-cli, you can deploy a new chatbot with just four lines of commands.


To install bothub:

$ pip install bothub

The bothub package works on python2 and 3 both.

Getting Started

You can build an echo chatbot simply by subclassing BaseBot class and overriding handle_message method.

 # -*- coding: utf-8 -*-

from __future__ import (absolute_import, division, print_function, unicode_literals)

from import BaseBot
from bothub_client.decorators import channel

class Bot(BaseBot):
    """Represent a Bot logic which interacts with a user.

    BaseBot superclass have methods belows:

    * Send message
      * self.send_message(message, chat_id=None, channel=None)
    * Data Storage
      * self.set_project_data(data)
      * self.get_project_data()
      * self.set_user_data(data, user_id=None, channel=None)
      * self.get_user_data(user_id=None, channel=None)
    * Channel Handler
      from bothub_client.decorators import channel
      def channel_handler(self, event, context):
        # Handle a specific channel message
    * Command Handler
      from bothub_client.decorators import command
      def command_handler(self, event, context, args):
          # Handle a command('/<command_name>')
    * Intent Handler
      from bothub_client.decorators import intent
      def intent_result_handler(self, event, context, answers):
          # Handle a intent result
          # answers is a dict and contains intent's input data
              "<intent slot id>" : <entered slot value>
    def default_handler(self, event, context):

When a bot receives a message from an user, it triggers handle_message method with event and context object.

An event is a dict which contains following items:

  • content: A message text received.

  • channel: Which channel (messenger platform) sent a message.

  • sender: Who sent a message. {"id": <user-id>, "name": "<username>}

  • chat_id: Chatroom ID where message is sent. It can be a 1:1 chatroom or group chatroom.

  • location: Location information if possible {"longitude": <float>, "latitude": <float>}

  • postback: A postback data.

  • new_joined: A boolean which indicates this bot was invited to some chatroom or not.

  • raw_data: A raw data itself messenger platforms sent.

You can respond to this message with various tools we provides.


To send a message, use a self.send_message method with a message you want to send.


In most cases, you may omit user_id and channel arguments. Then it replies to whom sent a message to your bot. Put values to those arguments when you want to specify a receiver.

You can send a message with rich controls like ‘quick replies’ or ‘buttons’ using Message object.

from bothub_client.messages import Message

message = Message(event).add_quick_reply('Go ahead')\
                        .add_quick_reply('Never mind')\
                        .set_text('May I reserve the seat?')

Message class provides these methods:

  • set_text(text)

  • add_url_button(text, url):

  • add_postback_button(text, payload)

  • add_quick_reply(text, payload=None, image_url=None)

  • add_location_request(text)

  • add_keyboard_button(text)


To store/retreive a property data, we provides following methods:

  • Project level

    • self.set_project_data(data): set data to a project

    • self.get_project_data(key=None): get data from a project

  • User level

    • self.set_user_data(data, user_id=None, channel=None): set user data

    • self.get_user_data(user_id=None, channel=None, key=None): get user data

data should be a dict. An existing properties not included in data will be ignored, not be deleted.

  • If user_id and channel is None, it regarded as a message sender.

  • When key is None, get whole dictionary will be returned. Otherwise, subtree of given key will be returned.

NLU Integeration

You can use nlu method to perform NLU after setup NLU integration at BotHub.Studio.

There are two styles to request to NLU service. (eg. to use

First, use event object to construct message and session_id.

def handle_message(self, event, context):
    response = self.nlu('apiai').ask(event=event)

Or, put explicit message and session_id by yourself.

def handle_message(self, event, context):
    response = self.nlu('apiai').ask(message='hello', session_id='customer1')

If you want to use a language other than english, use lang keyword argument on ask() function.

ask method returns a NluResponse object which contains attributes like:

  • raw_response: A raw response which NLU service returns.

  • action: A NluAction class object to identify intent and required parameters.

  • next_message: Next message text to respond NLU service recommend.

A NluAction object contains attributes like:

  • intent: Intent name

  • parameters: parameter dict

  • completed: A boolean indicates whether action completed

For incompleted action, you need to reply to user with next_message attribute of a NluResponse instance to complete action.


This package is licensed under AGPLv3 for non-commercial personal use. If you want to use this package for commercial use, please contact to

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

bothub-0.1.30.tar.gz (17.3 kB view hashes)

Uploaded source

Built Distribution

bothub-0.1.30-py2.py3-none-any.whl (22.9 kB view hashes)

Uploaded py2 py3

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page