Wordcloud for infosec hashtags
This script will create an image Wordcloud based on public posts in an instance's timeline. The general concept is to collect word clouds based on the hashtags of an instance of a public timeline of an instance. Trying to find new trends e.g. the polyfillio supplychain attack without reading all toots all the time.
It also generates an alttext proposal if you want to post the image online.
Example image
Running the code
This is designed to be run from a container. There is a prebuilt container image available.
podman run -v "$(pwd):$(pwd):z" -w "$(pwd)" quay.io/shivering-isles/mastocloud -m /app/shapes/mastodon.png -o example.png -i mastodon.example.com -t infosec
Build it yourself
You can build the image yourself using podman:
podman build -t quay.io/shivering-isles/mastocloud .
What are the options
| Command | Meaning | Example |
|---|---|---|
| -m | Masking image | masto.svg.png |
| -o | Output image | cloud.png |
| -i | Mastodon instance | mastodon.example.com |
| -t | hashtag to analyse | infosec |
Generic filtering
Add .txt file into the filters folder to remove hashtags from the wordcloud based on the hashtag that is queried. For example when using -t example the filters from filters/example.txt will be loaded, if the file exists.
The format is simple. One word per line. Lines starting with # or that are empty are ignored.
Credits
This project is based on mastocloud.