googlebot

Suggest that your team members check google before asking in a channel/message

The GoogleBot is a Slack bot that offers polite or not-so-polite responses to questions that should have been asked in a browser, not over Slack

Installation

As simple as installing any other global node package. Be sure to have npm and node (>= 4.3.2 version) installed and launch:

$ npm install -g slack-googlebot

Alternative Installation (binaries)

Starting at release 1.0.0, every release comes also with dependency-free binaries for linux, mac and windows that can be downloaded in the Releases page on GitHub.

Running the GoogleBot

To run the GoogleBot you must have a valid Slack BOT token to authenticate the bot on your slack organization. Once you get it (instructions on the next paragraph) you just have to run:

GOOGLEBOT_TOKEN=somesecretkey

Getting the BOT token for your Slack channel

To allow the GoogleBot to connect your Slack channel you must provide him a BOT token. To retrieve it you need to add a new Bot in your Slack organization by visiting the following url: https://yourorganization.slack.com/services/new/bot, where yourorganization must be substituted with the name of your organization (e.g. https://loige.slack.com/services/new/bot). Ensure you are logged to your Slack organization in your browser and you have the admin rights to add a new bot.

You will find your BOT token under the field API Token, copy it in a safe place and get ready to use it.

As an alternative you can create a bot by creating a custom application in the Slack developer portal. Inside the application settings you will be able to add a bot user and retrieve a OAUTH BOT token for it.

Configuration

The GoogleBot is configurable through environment variables. There are several variable available:

Installing on Heroku

A great place where to deploy our lovely GoogleBot is Heroku. We can go reasonably well with their free worker tier and the deploy process is reasonably easy and convenient. Let’s see how we can do that.

I am assuming you already have and account on Heroku and that you have installed and configured the Heroku toolbelt on your machine.

1. Create a new Heroku app

Create a local folder and install GoogleBot on it:

mkdir googlebot-myorg
cd googlebot-myorg
npm init -y
npm i --save slack-googlebot

Then create a new app on heroku:

heroku create googlebot-myorg

2. Configure your Heroku app:

heroku config:set --app googlebot-myorg GOOGLEBOT_TOKEN=xoxb-YOUR-AWESOME-BOT-TOKEN

(of course you need to replace xoxb-YOUR-AWESOME-BOT-TOKEN with your actual token).

You can add extra configuration by defining values for the other supported environment variables if you want to customize the behavior of the bot.

3. Initialize your Heroku app

Create an Heroku Procfile (service definition)

echo "worker: node_modules/.bin/googlebot" >> Procfile

Then prepare the project to be published through git:

git init
echo "node_modules/" >> .gitignore
git add --all
git commit -am "first version"
heroku git:remote --app googlebot-myorg

4. Publish and launch the Heroku app

Through heroku git

git push heroku master

Stop the web app (not present but started by default by Heroku) and run the worker:

heroku ps:scale web=0 worker=1

That's it, now go on your Slack organization and start enjoying GoogleBot!

Building the bot from source

If you downloaded the source code of the bot you can build the bot with

npm run build

Then you can run it with:

$ npm start

Don't forget to set your GOOGLE_TOKEN environment variable bedore doing so and to install all the dependencies (including dev ones with NPM or Yarn).

Bugs and improvements

If you find a bug or have an idea about how to improve the GoogleBot you can open an issue or submit a pull request, it will definitely make you a better person! :P

The Making of

Version 1 of GoogleBot has been developed based on the NorrisBot. A very detailed article has been published to explain every single line of code. It also explains you how to deploy the bot on a free Heroku instance, so you should give it a shot!

Enjoy your reading!

License

Licensed under MIT License. © Chris Rouffer.