grep: command-line tools #1

Welcome to part 1 of a blog series I’m going to do on command line tools, and how they’re useful as a bug bounty hunter. I’m going to highlight a different command line tool in each post, as well as do an extra post showing most of my bash aliases (there’s quite a few).

Disclaimer: This is mostly going to be use-cases for the tools related to how they are useful to me when bug hunting. I still rely on tools like ffuf, httprobe, aquatone, etc. but this series will cover how I use these command line tools to save time, process data, automate steps, and quickly extract meaningful conclusions.

Finding a specific thing

Anyone who has used grep knows it’s really great at finding a single string in a file or entire directory. My use-case is digging up some old endpoint or subdomain, or grepping for specific string like api_key. This is usually the command I use:

grep -Hrnia search_term

-H: show filename in output

-r: recursive in all subfolders

-n: show line number

-i: case insensitive

-a: process a binary file as if it were text

I remember those flags by thinking “hernia”. Pretty weird, but it works.

I also use the above command when grepping through source code because I pulled the source from all pages I was fuzzing with meg or the-od flag in ffuf (that’s the flag to save all responses into the (o)utput (d)irectory that you pass to the -od flag.)

The downside to this method is many lines are very long. If grep matches a string in a massive line (like minified javascript), it will fill up your terminal. For this, it’s a bit ugly looking, but I use the following:

grep -HrniaE ".{0,20}search_term.{0,20}"

The only flag I added was E which is the indicator to treat the search as a regular expression. The .{0,20} says to show 20 characters. Since I placed it before and after, it will return 20 characters before and after the search value, but nothing else. So this can make the output from your grepping a bit more readable if you’re grepping files with long lines.

As a filter

My second major use case for grep is filtering output from a tool. Sometimes it’s to get what I want. Sometimes it’s to exclude what I don’t want. For example, let’s say I’m using fff by tomnomnom to validate all the urls which I found by using Corben’s tool gau with a command such as this:

gau target.com | grep -vE "jpg|gif|png" | fff | grep " 200" > live.txt

gau takes a domain as input, and outputs links like:

https://target.com/js/main.js
https://target.com/login
https://target.com/photo.jpg

fff returns the url with a status, space delimited https://target.com/login 200

The first use of grep is filtering out any jpegs, gifs, or pngs. Those aren’t usually relevant for hunting. The -v is the flag that excludes matches. The -E is the flag which indicates the use of a regular expression. This allows me to use the OR operator (via the | pipe) to filter out a bunch of strings I want to exclude.

Then I used another grep filter to only keep the results that are 200 return codes. To be completely honest, this is only a simplified version of a line in my scanning script. I paired it down so that it was easier to understand. Here’s the full command for those of you who want to dig deeper.

while read line;do gau $line | grep -v -e '^$' | grep -vE "jpg|gif|png|css|jpeg|ttf|woff|svg" | tee -a gau.txt | fff --delay 100 | grep -E " 200| 403| 401| 405" | anew endpoints_live.txt;done < target.com.txt

Smaller use cases (more free #bugbountytips)

  • I use grep ? to filter links from gau and waybackurls which have params.
  • I use grep -v http if there’s a tool that returns full links and relative links like some uses of linkfinder and hakrawler and I deicde that I only want the relative links.
  • I use grep target.com as a filter when I’m worried the output from a crawling or scanning tool might pass something out of scope into another part of my script.
  • I use grep -v -e '^$' to remove empty lines from output. Probably a better way to do this.

Well, that’s about it. I hope you learned something. If you enjoyed this, you can tweet or follow me at https://twitter.com/rez0__ and/or subscribe to my newsletter

  • rez0