How I think about this site's security
What I want to protect on this site, the choices I've made, and the risks that still need attention.
Taking shape. Has structure and at least one real source or experiment.
Building this site gives me a chance to apply the questions I ask about other systems to something of my own. It’s a small website, but there are still things worth protecting and assumptions worth checking.
The main concerns are that nobody should be able to publish as me, the pages should stay available, and visiting shouldn’t involve unnecessary tracking. There are no user accounts, comments, form submissions, or database, which keeps the list of things to look after fairly short.
What the static setup helps with
The pages are generated ahead of time and served as files from a CDN. That removes the need for application sessions and a database, along with many of the problems that come with processing user input on a server.
It doesn’t make the site immune to security bugs. A compromised build can publish harmful files, browser-side code can have vulnerabilities, and someone with access to the hosting or domain account could change what visitors receive.
The areas I’m paying attention to are:
- Dependencies and the build process. The site relies on packages and CI, so reviewing dependencies, keeping a lockfile, and controlling what gets deployed matter even when the final pages are static.
- Accounts and domain settings. Access to the repository, hosting, and DNS is powerful. Strong account authentication, including hardware security keys, and carefully managed permissions belong in the deployment checks.
- What the browser is allowed to load. HTTPS and HSTS help protect connections, while the Content Security Policy restricts sources for scripts and other resources. Those controls have different jobs and aren’t a substitute for reviewing the code.
- Visitor privacy. I’ve kept analytics and third-party embeds out of the site. That reduces what the page asks a visitor’s browser to share, though it doesn’t mean the hosting infrastructure processes no request information.
What I’m still responsible for
Part of the reason for building the site is to learn from the implementation. Writing something myself makes me pay attention to decisions I might skim over when reading someone else’s code.
It also means I need to be willing to find mistakes in it. The mitigations above describe the approach I’m taking; the deployed settings and generated files still need checking.
If you notice something concerning, I’d appreciate a message through the site’s contact or security reporting details. Please keep any investigation non-disruptive so other people can continue using the garden.
This is a practical example of examining how the parts of a system fit together. Keeping the site small makes that easier, but I still have to do the examining.
Paths that lead here
- Thinking like an attacker · Threat-modeling this site asks similar questions from the defensive side.
Where this note points
- Thinking like an attacker · I'm interested in the gap between how a system is supposed to work and what its parts allow it to do.
More from these beds
- Feeling behind in cybersecurity · It's easy to feel behind when every new topic leads to several more. I'm trying to make my learning a little more manageable.
- Getting comfortable sharing my thoughts · I'm trying to make room for mistakes without letting the fear of them stop me from publishing.
- Getting into philosophy · A few questions that make philosophy feel relevant to everyday life, and the areas I'd like to understand better.
- How much time should I give a bug hunt? · I'm wondering whether the Kelly criterion can help me think about research time, and where that comparison breaks down.