Subdomain crawler
Brute force domain with a wordlist and return all the code 200 response.
Tech part
This script uses a number of open source projects to work properly:
- requests
- argparse
- python
Usage
usage: main.py [-h] [-d DOMAIN] [-w WORDLIST]
optional arguments:
-h, --help show this help message and exit
-d DOMAIN, --domain DOMAIN
Domain that you want to crawl
-w WORDLIST, --worlist WORDLIST
Wordlist default is in the working directory
python main.py -d google.com -w [default wordlist is in the repo]
Leave a Reply