DeGrabify

Provides an up-to-date list of Grabify domains, for use by adblockers.
The goal of this project is to reduce the risk of being doxxed by unknowingly clicking on IP logger links.
This project is not associated with Grabify.

This project is made up of two parts:

• A client which periodically retrieves a list of domains offered by the IP Logger service Grabify
• A webserver which provides this list of domains to adblock clients

Compatible with uBlock Origin, uBlacklist, and Hosts files
Easily self-hosted with Docker or compatible virtualization
Uses Flask, Python requests, and TinyDB

Setup

You can run this project in Docker or locally. Docker is recommended.

TLDR: docker pull, docker compose up. For a reverse proxy, forward the server through nginx/caddy/etc, get the forwarded URL. In the docker-compose.yml, set the base proxied domain, set the proxy level -p, disable the exposed port, and connect the server to your bridge network.

Docker

Server

• Clone this repository
  • git clone https://github.com/JonasLong/DeGrabify
  • cd DeGrabify
• Consider using a reverse proxy as described below
• Configure the container in docker-compose.yml
• docker compose up
• Continue to the Client section

Updating

• git fetch
• git pull
  • You may need to git stash if you've made changes to the config, then merge your stashed changes into main
• docker compose up --build

Run behind a reverse proxy

• Change the -p value in the commands section of the server config from 0 to 1
• Uncomment both networks: sections in the compose file. Change the name: default value to the name of your bridge network
  • This will ensure that the server is connected to the same external bridge network that your reverse proxy is running on.
• Comment out the ports: section
• Change the -s option in server/server.py from http://127.0.0.1:5000 to the base domain your server will be hosted on (e.g. https://degrab.example.com). If this is set incorrectly, the "Subscribe" buttons won't work but nothing else will be affected.
• Configure your reverse proxy
  • scheme: http
  • domain: degrabify-server-1
  • port: 5000

Development

• Clone this repository
  • git clone https://github.com/JonasLong/DeGrabify
  • cd DeGrabify
• Run like a normal docker container, but use docker compose up --build to rebuild any file changes

Client

To install with your adblock or Hosts file:

• Navigate to the webserver address in a browser
• Click the relevant subscribe link, or follow the instructions in uBlockOrigin-HUGE-AI-Blocklist for the provided URLs.
• Most likely you'll go to uBlock or uBlacklist and import a filter, passing it the URL
• Don't copy-paste the list of domains into your filter list- it won't auto-update as the grabify domains change

Run locally (no Docker)

• Install python and the pip packages flask and tinydb
• Clone this repository
  • git clone https://github.com/JonasLong/DeGrabify
  • cd DeGrabify
• In the console, run
  • python crawler/crawler.py -d database/sites.json
  • python server/server.py -d database/sites.json
• If you'd like crawler.py to run on a schedule, install cron and run
  • chmod 700 crawler/cron-install.sh
  • ./crawler/cron-install.sh $(pwd)/crawler.py '0 12 * * *' -d $(pwd)/database/sites.json
    • The cronjob will be written to /etc/cron.d/crawl-cron. Logs will be saved in /var/log/crawl.log.
  • cron
• Open http://127.0.0.1:5000 in a browser

Roadmap

• Abstract list formatting into the config/.env
• Make a python requirements file
• Make a ghcr build
• Double check header specs for ublock and ublacklist
  • any missing fields?
  • figure out if "access time" field ruins caching
• Use Flask in production
• Pretty up the homepage
• Better disguise crawler
• See if the "r" param in the URL changes over time as a revision #
• Better naming for the containers(?)

Acknowledgements

Inspired by the following projects, check them out as well:

• https://github.com/TMAFE/anti-grabify
• https://github.com/0n1cOn3/The-AntiIPGrabber-Blocklist
• https://github.com/laylavish/uBlockOrigin-HUGE-AI-Blocklist
• https://github.com/NotaInutilis/Super-SEO-Spam-Suppressor