markov-bot

Quick Twitter bot using Markov chains and AWS Lambda. Using the downloader script, you can download a dataset of tweets from a given account. The bot will tweet a generated tweet every 4 hours.

Live example: @imalsoblackbear

Requirements

• Node.js
• Python
• aws-cli: Install via Amazon's guide
• Serverless: npm install -g serverless

Setup

1. Run the following in the command line:

git clone https://github.com/afflitto/markov-bot && cd markov-bot/bot
npm install
cp config.js.template config.js

cd ../downloader
sudo pip install tweepy
cp config.py.templage config.py

2. (optional) Modify serverless.yml

• You may want to change the service name if you plan on running more than one bot.
• You can change the rate to determine how often the bot will run.

3. Create a Twitter app on developer.twitter.com

• You probably want to create a new Twitter account and apply to be a developer (otherwise the bot will tweet from your account). Then you can create an app and generate consumer and access tokens.
• Fill in config.js and config.py with your app's consumer and access token keys/secrets.

4. Download tweets: python downloader.py {username}, filling in {username} with a twitter username.

5. Copy the tweets to the bot directory: cp tweets*.json ../bot/tweets.json

6. Deploy: cd ../bot && serverless deploy

The twitter bot will automatically run every 4 hours (or whatever you changed the rate to in serverless.yml). To run it manually, run serverless invoke -f tweet.