All Projects → primaryobjects → Chatskills

primaryobjects / Chatskills

Run and debug Alexa skills on the command-line. Create bots. Run them in Slack. Run them anywhere!

Programming Languages

javascript
184084 projects - #8 most used programming language

Projects that are alternatives of or similar to Chatskills

Awesome Bots
The most awesome list about bots ⭐️🤖
Stars: ✭ 2,864 (+1574.85%)
Mutual labels:  chatbot, alexa, chatbots, slack-bot, slackbot, slack
alexa-ruby
Ruby toolkit for Amazon Alexa service
Stars: ✭ 17 (-90.06%)
Mutual labels:  alexa, amazon, amazon-echo, alexa-skill, amazon-alexa
Amazon Alexa Php
Php library for amazon echo (alexa) skill development.
Stars: ✭ 93 (-45.61%)
Mutual labels:  alexa, echo, alexa-skill, amazon-echo, amazon
Alexa Skill Kit
Library for effortless Alexa Skill development with AWS Lambda
Stars: ✭ 278 (+62.57%)
Mutual labels:  chatbot, alexa, alexa-skill, amazon-echo, amazon
cookiecutter-flask-ask
Cookiecutter template for Alexa skills based on the fantastic Flask-Ask framework 🍾🗣❓
Stars: ✭ 51 (-70.18%)
Mutual labels:  alexa, amazon-echo, alexa-skill, amazon-alexa
Sactive Bot
😈 An extensible chat bot framework. sactive-bot is an evolution of the open source hubot project. - https://www.shipengqi.top/sactive-bot .
Stars: ✭ 212 (+23.98%)
Mutual labels:  chatbot, slack-bot, slackbot, slack
Alexa Skills Kit Sdk For Java
The Alexa Skills Kit SDK for Java helps you get a skill up and running quickly, letting you focus on skill logic instead of boilerplate code.
Stars: ✭ 758 (+343.27%)
Mutual labels:  alexa, amazon-alexa, amazon-echo, amazon
Botonomous
A PHP Framework For Creating Autonomous Slack Bots
Stars: ✭ 109 (-36.26%)
Mutual labels:  chatbot, slack-bot, slackbot, slack
Awesome Amazon Alexa
🗣Curated list of awesome resources for the Amazon Alexa platform.
Stars: ✭ 458 (+167.84%)
Mutual labels:  alexa, alexa-skill, amazon-alexa, amazon-echo
Fantasy football chat bot
GroupMe Discord and Slack Chatbot for ESPN Fantasy Football
Stars: ✭ 166 (-2.92%)
Mutual labels:  chatbot, slack-bot, slackbot, slack
Slacker
Slack Bot Framework
Stars: ✭ 495 (+189.47%)
Mutual labels:  chatbot, slack-bot, slackbot, slack
Voicewp
Create Alexa Skills through WordPress
Stars: ✭ 132 (-22.81%)
Mutual labels:  alexa, alexa-skill, amazon-alexa, amazon
Obed Bot
🍴 Obed Slack Bot, na požiadanie kontroluje aktuálnu ponuku denného menu reštaurácii. [only for slovak|czech users]
Stars: ✭ 32 (-81.29%)
Mutual labels:  slack-bot, slackbot, slack
Slack Ruby Client
A Ruby and command-line client for the Slack Web, Real Time Messaging and Event APIs.
Stars: ✭ 957 (+459.65%)
Mutual labels:  slack-bot, slackbot, slack
Awesome Voice Apps
🕶 A curated list of awesome voice projects, tools, and resources for Amazon Alexa, Google Assistant, and more.
Stars: ✭ 138 (-19.3%)
Mutual labels:  alexa, alexa-skill, amazon-alexa
Botpress
🤖 Dev tools to reliably understand text and automate conversations. Built-in NLU. Connect & deploy on any messaging channel (Slack, MS Teams, website, Telegram, etc).
Stars: ✭ 9,486 (+5447.37%)
Mutual labels:  chatbot, chatbots, conversational-ui
Wc2018 Slack Bot
World Cup 2018 Slack Bot
Stars: ✭ 11 (-93.57%)
Mutual labels:  slack-bot, slackbot, slack
Jbot
Make Slack and Facebook Bots in Java.
Stars: ✭ 1,148 (+571.35%)
Mutual labels:  slack-bot, slackbot, slack
Claudia Bot Builder
Create chat bots for Facebook Messenger, Slack, Amazon Alexa, Skype, Telegram, Viber, Line, GroupMe, Kik and Twilio and deploy to AWS Lambda in minutes
Stars: ✭ 1,717 (+904.09%)
Mutual labels:  chatbot, alexa, slack
Jovo Framework
🔈 The Open Source Voice Layer: Build Voice Experiences for Alexa, Google Assistant, Samsung Bixby, Web Apps, and much more
Stars: ✭ 1,320 (+671.93%)
Mutual labels:  alexa, alexa-skill, amazon-alexa

Chatskills

Run Alexa apps on the command-line. Run them in Slack. Run them anywhere! Supports Amazon Alexa skills and intents.

$ npm install chatskills

Chatskills is a quick and easy way to run Alexa apps outside of Amazon. Easily create your skills and intents and run them right on the command-line!

Chatskills does not require a server and can run directly in the console. It can also run on the web, or Slack, or anywhere. It handles requests from multiple users and maintains session memory. When a user starts a conversation with one of the skills, the skill continues to execute within a session context, until the skill terminates.

Here's what an Amazon Alexa app looks like, running on the command-line.

Example

> chatskills, ask hello to say hi.
Hello, World!
> chatskills, ask horoscope for Scorpio.
Things are looking up today for Scorpio.
> chatskills, ask funny to tell me a joke
Knock knock.
> who's there?
Banana.
> banana who
Knock knock.
> whos there
Orange.
> orange who?
Orange you glad I didn't say banana?

In this example, the user accesses three different skills: hello, horoscope, and funny.

Usage

Using chatskills is easy. Use the Alexa syntax to add a new skill, then create some intents. Here's a simple example.

var chatskills = require('chatskills');

// Create a skill.
var hello = chatskills.app('hello');

// Create an intent.
hello.intent('helloWorld', {
    'slots': {},
    'utterances': [ '{to |}{say|speak|tell me} {hi|hello|howdy|hi there|hiya|hi ya|hey|hay|heya}' ]
    },
    function(req, res) {
        res.say('Hello, World!');
    }
);

// Respond to input.
chatskills.respond('chatskills, ask hello to say hi', function(response) {
    console.log(response);
});

In the above example, the utterances grammar automatically expands to match on the following phrases:

helloWorld      to say hi
helloWorld      say hi
helloWorld      to speak hi
helloWorld      speak hi
helloWorld      to tell me hi
helloWorld      tell me hi
helloWorld      to say hello
helloWorld      say hello
helloWorld      to speak hello
helloWorld      speak hello
helloWorld      to tell me hello
helloWorld      tell me hello
helloWorld      to say howdy
...

To interact with the chatbot using this skill, say any of the target phrases. In the above example, we've used the phrase "to say hi", but you can match against any of the generated phrases. For example:

> chatskills, ask hello to tell me hi
Hello, World!
> chatskills, ask hello to say hello
Hello, World!
> chatskills, ask hello to say howdy
Hello, World!

Reading from the Console

To create a chatbot that runs locally on the console, just include a loop for reading input.

var readlineSync = require('readline-sync');

// Console client.
var text = ' ';
while (text.length > 0 && text != 'quit') {
    text = readlineSync.question('> ');

    // Respond to input.
    chatskills.respond(text, function(response) {
        console.log(response);
    });
}

If you're using async calls in your skills (such as request, etc) then you'll want to use an async loop, instead of the while loop above. Here's an example.

Reading from Slack

You don't have to use just the console! You can run your chatbot anywhere, like Slack. See here for full example.

var SlackBot = require('slackbots');
var bot = new SlackBot({ token: token, name: 'awesome' });

// Listen to slack messages.
bot.on('message', function(message) {
    // Reply to humans.
    if (message.type == 'message' && message.text && message.subtype != 'bot_message') {
        var author = getUserById(message.user);
        var channel = getChannelById(message.channel);

        // Respond to input, use author.name as the session id.
        chatskills.respond(message.text, author.name, function(response) {
            if (channel) {
                // Public channel message.
                bot.postMessageToChannel(channel.name, response);
            }
            else {
                // Private message.
                bot.postMessageToUser(author.name, response);
            }
        });
    }
});

Chatskills running on Slack

Creating a Skill

Skills are programs that your chatbot can run. They consist of intents, which are composed of utterances (phrases to match from the user input), responses, and session memory. Each skill can access session memory, so you can store and retrieve variables to help with responding intelligently to the user.

Here's an example of creating a new skill, named "horoscope".

var horoscope = chatskills.app('horoscope');

Creating an Intent

Skills are made up of intents. This is where input from the user is matched against an array of utterances. When a match is found, that intent is executed. An intent can get/set variables in the user session by calling req.get('variable') and req.set('variable', value). An intent can output a response by calling res.say('hello').

Here's an example of creating a new intent for the skill "horoscope".

horoscope.intent('predict', {
    'slots': {'SIGN':'LITERAL'},
    'utterances': [ 'for {signs|SIGN}' ]
    },
    function(req, res) {
        res.say('Things are looking up today for ' + req.get('SIGN') + '.');
    }
);

This intent can be interacted with like this:

> chatskills, ask horoscope for Scorpio
Things are looking up today for Scorpio.

Launching a Skill

There are two ways to begin running a skill.

Using an Intent to "Run"

The first way to launch a skill is to create an intent such as, "run". This would let you enter: "chatskills, ask [skillname] to run.". Provided the intent has a return value of true (to keep the session alive), your skill will now be running.

An example of a "run" skill can be found in guessinggame.

app.intent('run', {
        "slots": {},
        "utterances": ["{to|} {run|start|go|launch}"]
    }, function(req, res) {
        var prompt = "Guess a number between 1 and 100!";
        res.say(prompt).reprompt(prompt).shouldEndSession(false);
    }
);

Using the Launch Method

The second way to launch a skill is to create a launch method to automatically run upon starting your app. Then simply call chatskills.launch(app) to start your skill. You can pass the skill or the name of the skill. You can also provide an optional unique sessionId.

Example: chatskills.launch(app) or chatskills.launch('horoscope') or chatskills.launch('horoscope', 'some-unique-id').

Here's a complete example.

var chatskills = require('./lib/chatskills');
var readlineSync = require('readline-sync');

// Create a skill.
var hello = chatskills.app('hello');

// Launch method to run at startup.
hello.launch(function(req,res) {
    res.say("Ask me to say hi!");

    // Keep session open.
    res.shouldEndSession(false);
});

// Create an intent.
hello.intent('helloWorld', {
    'slots': {},
    'utterances': [ '{to |}{say|speak|tell me} {hi|hello|howdy|hi there|hiya|hi ya|hey|hay|heya}' ]
    },
    function(req, res) {
        res.say('Hello, World!');
    }
);

// Start running our skill.
chatskills.launch(hello);

// Console client.
var text = ' ';
while (text.length > 0 && text != 'quit') {
    text = readlineSync.question('> ');

    // Respond to input.
    chatskills.respond(text, function(response) {
        console.log(response);
    });
}

Starting and Ending a Session

When a user provides input, the input is matched against each skill and their list of intents. When a match is found, a new session starts, and the skill begins executing.

When a session has started for a user, the activated skill's intent can get/set variable values within the session. This allows you to store and retrieve data.

While a session is open for a user, all input from the user is directed to the activated skill. In this manner, the user does not need to re-request a skill ("chatskills, ask hello to say hi"). Instead, the user can simply provide text, which will be matched against the currently executing skill's intents.

An intent can keep a session open by returning true or by calling res.shouldEndSession(false) and end a session by returning false or by calling res.shouldEndSession(true). An intent may also omit a return statement, which is the same as returning false.

For an example using session, see the horoscope skill. Notice, the intent asks the user a question and then returns true to keep the session going. The intent only returns false once a valid response is given, thus, ending the session.

In summary, when a user session is open, all input from the user is directed to the skill. When a user session is ended, input from the user must be received in the format, "chatskills, ask [SKILL] text", to execute a new skill.

Session Timeout

The default session timeout is 1 hour of no input from the user. To change the session timeout, set chatskills.timeout = 3600, where the value is specified in seconds. To disable session timeout, set the value to 0.

Changing the Chatbot Name

The default chatbot name is "chatskills". All requests to execute a skill must begin with the chatbot name. For example, "chatskills, ask hello to say hi". To customize the chatbot name, use the following:

chatskills.name('awesome');

Verbose Output

To display warnings and errors, set chatskills.verbose = true.

Schema and Utterances

Chatskills uses alexa-app to generate many sample utterances from your intents. For a more detailed description of utterances, see here.

Schema Syntax

Pass an object with two properties: slots and utterances.

app.intent('sampleIntent',
    {
        "slots":{"NAME":"LITERAL","AGE":"NUMBER"}, 
        "utterances":[ "my {name is|name's} {names|NAME} and {I am|I'm} {1-100|AGE}{ years old|}" ]
    },
    function(request,response) { ... }
);

Slots

The slots object is a simple Name:Type mapping. The type must be one of Amazon's supported slot types: LITERAL, NUMBER, DATE, TIME, DURATION

Custom Slot Types

As a replacement for the LITERAL slot type, which is no longer being supported by Amazon, it is recommended to use custom slot types in its place. Here is an example of defining a custom slot type for DragonType.

app.intent('attack', 
    {
        'slots': { 'DragonType': 'DRAGONTYPE' },
        'utterances': ['{attack|fight|hit|use} {sword|dagger|wand} on {-|DragonType} dragon']
    }, function(request,response) {
        response.say('You are attacking the ' + request.slot('DragonType') + ' dragon!');
    }
);

You can include custom slot types within utterances by using the syntax {-|CustomTypeName}. This indicates that the term should come from a list of values for the custom slot type. In the example above, the utterance uses the term {-|DragonType}, indicating a term should come from the list of values (shown below). For chatskills, a list of values does not need to be provided - any word will be accepted for a custom slot type and used as its value.

If publishing to the Amazon Alexa service, you would provide the custom slot types for DragonType by specifying the type name and a list of values. For example:

Type: DRAGONTYPE

Values:

golden
fire
ice
water
snow

Note, chatskills and Amazon Alexa will actually accept any word for the custom slot value. It doesn't have to match a word from the list of values. In this manner, custom slot types are similar to LITERAL.

Utterances

The utterances syntax allows you to generate many (hundreds or even thousands) of sample utterances using just a few samples that get auto-expanded. Any number of sample utterances may be passed in the utterances array. Below are some sample utterances macros and what they will be expanded to.

Multiple Options mapped to a Slot

"my favorite color is {red|green|blue|NAME}"
=>
"my favorite color is {red|NAME}"
"my favorite color is {green|NAME}"
"my favorite color is {blue|NAME}"

Generate Multiple Versions of Static Text

This lets you define multiple ways to say a phrase, but combined into a single sample utterance

"{what is the|what's the|check the} status"
=>
"what is the status"
"what's the status"
"check the status"

Auto-Generated Number Ranges

When capturing a numeric slot value, it's helpful to generate many sample utterances with different number values

"buy {2-5|NUMBER} items"
=>
"buy {two|NUMBER} items"
"buy {three|NUMBER} items"
"buy {four|NUMBER} items"
"buy {five|NUMBER} items"

Number ranges can also increment in steps

"buy {5-20 by 5|NUMBER} items"
=>
"buy {five|NUMBER} items"
"buy {ten|NUMBER} items"
"buy {fifteen|NUMBER} items"
"buy {twenty|NUMBER} items"

Optional Words

"what is your {favorite |}color"
=>
"what is your color"
"what is your favorite color"

Using a Dictionary

Several intents may use the same list of possible values, so you want to define them in one place, not in each intent schema. Use the app's dictionary.

app.dictionary = {"colors":["red","green","blue"]};
...
"my favorite color is {colors|FAVEORITE_COLOR}"
"I like {colors|COLOR}"

Displaying Home Cards

You can display Amazon Alexa Home Cards by handling the card object returned in the response method. When using alexa-app, the home card will be displayed in the Amazon Alexa App on your mobile device. When using chatskills, the home card can be handled in the chatskills.respond() callback method, which returns two arguments: response and card.

Using the card object, you can display the card's text and image in any manner you wish. For example, if hosting your chatskills app in Slack, you may want to show the image as embedded media. Likewise, if hosting as a text chatbot on the console, you may simply want to output the card as text.

Below is an example.

app.intent('example', {
        "slots": {},
        "utterances": ["show a card"]
    }, function(req, res) {
      // Show home card in Alexa app.
      res.card({
        type: 'Standard',
        title: 'My Awesome Card', // this is not required for type Simple or Standard 
        text: 'This is an example of an Alexa home card.',
        image: { // image is optional 
          smallImageUrl: 'http://www.yoursite.com/image.jpg', // required 
          largeImageUrl: null
        }
      });
});

// Respond to input.
chatskills.respond(text, function(response, card) {
    if (!card) {
      // Text response from res.say() method.
      console.log(response);
    }
    else {
      // Home card response from res.card() method.
      console.log('[DISPLAYING CARD: Title=' + card.title + ', Text=' + card.text + ']');
    }
});

License

MIT

Author

Kory Becker http://www.primaryobjects.com/kory-becker

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].