there was previous version but i modify it to be petter
A Python script to extract subdomains for a given domain using the crt.sh website. This script utilizes requests
, BeautifulSoup
, argparse
, and colorama
libraries to fetch and parse the data and provide a user-friendly command-line interface with colored output.
-
Extract subdomains for a given domain using crt.sh
-
Save extracted subdomains to a text file (optional)
-
Colored terminal output for better readability
Before running the script, you need to install the required libraries. You can do this by running:
pip install requests beautifulsoup4 colorama
To use the script, run the following command in your terminal:
python crtHunter.py -d DOMAIN [-o OUTPUT_FILE]
Or you can add list :
python crtHunter.py -l DOMAINS_LIST [-o OUTPUT_FILE]
-
Replace
DOMAIN
with the domain you want to search for subdomains. -
The
-o OUTPUT_FILE
flag is optional. If provided, the script will save the extracted subdomains to the specifiedOUTPUT_FILE
.
python crtHunter.py -d example.com -o subdomains.txt
Or use list of domains:
python crtHunter.py -l domains_lit.txt -o subdomains.txt
This command will extract subdomains for example.com
and save them to a file named subdomains.txt
.
Or extract all the subdomains for each domain in the domains_list.txt
...