Here's another "meta" long-form hacking tip that has paid its weight in gold.
== Don't rely on TOO much automation ==
A thread 🧵
🚨follow, retweet, & like for more 🚨
Some examples:
👇
1/x
In Recon:
Let's start with subdomain enumeration techniques.
Tools like Amass & Subfinder are just tools using web API's & scraping to pull subdomains from datasets on the internet...
2/x 👇
However, it's been shown live by many hackers (like @NahamSec) that working with a dataset or website directly, like cert.sh , can find nested subdomains, or more results than a tool.
Why does this happen? Parsing is hard, rate limits exist, etc.
3/x 👇
Another example in recon, Screenshotting:
Using a screenshot tool on live web servers can miss things due to timing, complex redirects, cert issues, cloud/CDN configs, etc.
Many testers notice this right away if they ever switch to just opening URLs in browser.
JavaScript parsing is another example. LinkFinder or derivatives give:
Full URLs
Absolute URLs or dotted URLs
Relative URLs with at least one slash
Relative URLs without a slash
But will not give you anything for minified or obfuscated js
6/x 👇
Finally, and known by most: Vuln Scanners.
Even dynamic ones are just throwing injection strings against parameters and routes. They then parse the page returned looking for a set of conditions and if X condition(s) are met (usually a regex) then they alert you...
7/x 👇
What you miss out on is the context around the fuzzing:
the error codes/text
the return time
the content size
etc
It's all invisible to you in most cmdline tools and each of those could lead you to a juicy bug.
8/x 👇
So, throw your automation out the window right?!
No 😅
Use your intuition/judgment when testing. Know what your tools are doing & what they can miss. Learn to love manual testing, slow & deep testing can yield great returns.
🚨follow, retweet, & like for hacker tips🚨
9/x
• • •
Missing some Tweet in this thread? You can try to
force a refresh
How to find assets no other bug hunters have found.
One of my simple "secrets" for years.
Little automation exists for it.
💸💸💸
a thread🧵
🚨follow, retweet, & like for more hacker tips!🚨
1/x
When approaching a bounty, the scope is important. Not only the domain list but, all the text.
There are about ~30 paid bounty programs across the major platforms that are explicitly open scope or have the wording right under the scope section that says something like...
2/x
"If you find anything else that you believe to belong to XYZ company, report it and we will assess its validity. It may not result in a bounty"
But.. To be honest, criticals usually DO get paid.
3/x
The Web Application Hacker's Handbook is a pre-requisite for all web assessments. Do not sleep on it due to publish date. It remains the 👑 book for web assessment.
Every time you find a bug, invest time upfront to write up a REALLY great submission template. This includes impact assessment and remediation advice. Then re-use it for the rest of your career.
Mistake Two:
I'll stop hacking
Often, on a bug bounty, I'll submit something good and stop and wait around for a bit to see how the client responds.
You should always have a backup program to analyze while you hack on a new program.
How I hacked access to the most sensitive areas of a porn site using only low severity vulnerabilities.
Here's how I did it...
👇
🚨follow, retweet, & like for more hacker stories!🚨
1/x
I was once contracted to do a penetration test on a porn site.
This site was more than your average view-only site. It had community functions to:
- share images privately with other members
- had private paid cam access
- DMing
- and a store for sexy gifts!
2/x 👇
I started with normal usage of the site, registering my own account on each of the websites.
The goal set by the client was to access a restricted set of photos in a certain account, plus notify them of any other security vulnerabilities I found.
I'm looking for people to jump in and give me their perspectives. This is all speculative and in US hyper inflated markets.
A Sr/Principle Security Tester in the US can command $150-200k salary in big markets (SFO, LA, NY).
👇1/x
That's 260 workdays.
$770 US a day.
$22k a month. Pre-tax.
That means as a FT Bug Hunter you need to come close to those numbers.
Now let's look at programs...
👇2/x
(napkin math)
The avg payout across all paid programs seems to come in at $500. That doesn't bode well for how impact is rated or how achievable a critical/P1 is for the bounty hunter.
On the high end, if you do find a Crit/P1, the average is $10k across big brands.