Bug hunter's search for meaning in Low/Medium bugs.


Three years ago, I started to find bugs more often. 95% of my bugs were low/medium. Although bounties were nice, I often wondered if my reports have any impact on the security of companies I submitted to. If my reports were useful to developers, security teams, or they were a waste of time, and bug bounty programs have to accept low/medium just to comply with bug bounty community standard, but what they really want are high/crit reports.


I felt unease, while writing a new report for a low bug. I often thought If this report was really needed, or it was a waste of time for program staffs and everyone involved. Of course, the bounties were still larger than these doubts. So I reported.

I tried to answer these doubts many times, some answers I had come up with are:

  • Many of my low bugs are fixed, that means developers care about these bugs. Otherwise, they can just leave them in backlog for years.
  • They have option to not pay for low bugs at all, but they pay, that means these reports have value to them.
  • 7 low/medium bugs attack individually can cause many small integral issues on target websites. Although the impact are small, users can feel the lack of integrity here and there, which makes them feel unease using the website. Also, it is hard for developers to identify and recover all these integral issues.
  • If these small holes are not fixed, they will start the Broken Window Effect and start to drag the security quality of the whole website down.

After these answers, my doubts was smaller, but not completely gone. Fortunately, I begin to see a better answer for my doubts. It is in a website that I have reported around 120 valid bugs on in the past 3 years, most of them are low/medium.

When we test on a website for a few hours, we begin to get a feeling of how good the security stance of the website is. If it feels solid? When we enter unexpected data, the website responds in a uniform way or weird way? If it feels solid, I will likely switch target after a few days. So do attackers.

Three years ago, this website (the 120 bugs one) has small data leak and trivial integral issues everywhere, many of them are not impactful enough to even be a low bug. But that's a good sign for me to spend more time here. So do attackers.

Most of my bugs in this website are fixed (112). Not only that, I can see developers of this website are getting better at writing secure code, such as, reuse existing APIs for fetch/edit data in new features when possible, instead of writing new APIs. They also consider many business logic cases. They introduce new security rules that breaks many attack chains, while still keep the UX/UI harmonious. I can see some of these rules are inspired by my low/medium reports.

Now testing on this website feels like hitting a medieval castle stone wall. Very solid. All holes are closed, even small ones.

I am glad that I endured through my doubts and finally see the fruitful impact of my "trivial" reports years ago.




Comments

Popular posts from this blog

Beginner Tutorial - How to learn the Technical Skill and Hacker Mindset That Are Required to Find Your First Bug Bounty.

The power of focus

How to succeed in bug bounty as a non-talented bug hunter