Show HN: Osint tool that finds exposed files on domains
Comments
phoronixrly
bblb
I need to protect against the malicious good guys, the shodans of the world. Peeping inside your windows and trying your front door handles. Querying every string imaginable and port pinging all 65536 of them.
And I need to protect against the actual criminals already inside my house looking for something to steal. Scouting out the easy target to setup their ransomwares.
graemep
Lots of crawlers do this. I have never seen a webserver that does not get a variety of these from obviously different sources. Even just an ssh port will get a lot of malicious login attempts.
phoronixrly
I don't get what you mean by that. Everyone does it so it's OK?
No - scanning without consent is abuse and anyone doing this is malicious regardless of what they tell themselves or anyone else.
graemep
I am not saying its OK. I am saying it is probably not this crawler that is hammering your site, and definitely not only this crawler. I specifically mentioned malicious attempts at ssh logins.
technion
There's an astounding amount of .DS_Store showing up - I hadn't realised how common it apparently is for people to accidentally upload this.
stingraycharles
It’s a terrible design from Apple to expose this metadata like this, it’s one of my biggest pet peeves.
graemep
KDE does something similar (but not as bad?) with .directory files.
marysol5
CT logs are funny, not enough people know about them. Whenever I speak to people they're shocked to know that "internal" subdomains get exposed through them.
seethishat
Wildcard certs were somewhat useful for hiding this. Not sure that's still true.
keepamovin
In the early days of the web you could do a search on google like
path:/etc/passwd
AndReics
i thought it was still possible!
Luckily security has come a long way, but as shown by the project of OP, we are not quite there yet.
katemaster009
I remember seeing examples like this in security courses.
It was always surprising how many servers accidentally exposed sensitive files.
keepamovin
Yes. Apache misconfigurations were a big one iirc. There were also basic auth files, databases and probably classified/proprietary information.
Similar to how the telephone network used to have all kinds of unsecured entry points that people explored, leading to business phone systems, strange modems, and even international “trunk” lines and operator capabilities.
It was a wild frontier
Elliott-Diy
Is this just re-skinned leakix.net? One of my honeypots for them is showing the same results.
PatchRequest
Nope, i crawl everything myself.
sandeepkd
Its interesting and not interesting at the same time based on some of the search results
Almost all of them seem like home projects being deployed with ease in mind than security. The common thread seems to be the fact that most of them are phishing website, not sure if thats a business model to target here?
Avery29
Nice tool. I’d like to understand what kinds of businesses the customers using this website are in.
sub-e12
Awesome tool, fast and right to the point!
cvadict
searching for .gov reveals 0 matches... doubt
PatchRequest
I manually blacklisted .gov from being crawled on my side. felt like it wasn't worth the potential trouble
marysol5
Meanwhile you crawl anybody else?
.gov isn't magically any more able to take you to court than a private org
sandeepkd
My guess is that they ran selective search on the domains which get registered with any registrar, thats the trigger to start the search. .gov domains are not managed by your typical registrar which is selling the domain registration information to all these downstream partners/scavengers (for lack of better word)
jadamson
The OP says it's using CT logs, not new domain registrations. The approach you have in mind would not include subdomains and would be less likely to coincide with a new server being configured.
sandeepkd
Yes CT is explicitly stated source which is why I qualified it with "Guess" for domain registration. There were couple reasons for that -
1. Quite a few websites in the search results where just on HTTP
2. The .gov sites do use public certificate authorities like digicert, verisign, amazon & letencrypt so they would have been captured unless they are removed explicitly
And yes the domain registration would not include subdomains
marysol5
.gov.uk plenty of them use standard CA's
manchester.gov.uk uses LE
So is this the crawler that has been constantly hammering all my applications searching for these files from the very second I first issue a TLS cert for them? Thanks to you I've had to put fail2ban on all my public-facing web servers...
How about you be a good netizen and make it so people can request to be scanned and don't proactively do it, let alone constantly keep hammering them with requests?