Podcast Detail

SANS Stormcast Monday, July 28th, 2025: Linux Namespaces; UI Automation Abuse; Autoswagger

If you are not able to play the podcast using the player below: Use this direct link to the audio file: https://traffic.libsyn.com/securitypodcast/9544.mp3

Podcast Logo
Linux Namespaces; UI Automation Abuse; Autoswagger
00:00

Linux Namespaces
Linux namespaces can be used to control networking features on a process-by-process basis. This is useful when trying to present a different network environment to a process being analysed.
https://isc.sans.edu/diary/Sinkholing%20Suspicious%20Scripts%20or%20Executables%20on%20Linux/32144

Coyote in the Wild: First-Ever Malware That Abuses UI Automation
Akamai identified malware that takes advantage of Microsoft’s UI Automation Framework to programatically interact with the user’s system and steal credentials.
https://www.akamai.com/blog/security-research/active-exploitation-coyote-malware-first-ui-automation-abuse-in-the-wild

Testing REST APIs with Autoswagger
The tool Autoswagger can be used to automate the testing of REST APIs following the OpenAPI/Swagger standard.
https://github.com/intruder-io/autoswagger/

Podcast Transcript

 Hello and welcome to the Monday, July 28, 2025 edition
 of the SANS and the Storm Center's Stormcast. My name is
 Johannes Ulrich, recording today from Jacksonville,
 Florida. And this episode is brought to you by the SANS.edu
 graduate certificate program in cloud security. In diaries
 this weekend, we have one by Xavier looking into, well,
 we'll need Linux feature. And Xavier looks at it from sort
 of a reverse analysis point of view, but it's really very
 applicable for a number of different security tasks. And
 that's Linux namespaces. Essentially, each process in
 Linux may have sort of its own namespace, its own view of the
 environment. And in this particular case, Xavier looked
 at networking, where first of all, you are able to just
 simply turn off networking capabilities for a particular
 process with the sudo unshare dash dash net bash command.
 That basically gives you then a bash shell without
 networking. And now if you try to analyze some malware, well,
 that malware can no longer communicate outbound. But it
 goes more fine grain than that. You can also just set up
 a different routing table for this particular process. And
 for example, redirect traffic to sinkholes and the like.
 Quite often when you're analyzing malware, you don't
 want to turn off networking altogether because the malware
 will not run if it can't download second stages and
 such. But you just want to capture like what is that
 second stage it's downloading and then send the request to a
 sinkhole where you're just recording the HTTP requests.
 And that's sort of where this feature is really helpful. But
 like I said, namespaces in Linux can do a lot more
 things. There's file systems and mounts and similar
 features that you have available as you have for
 networking. And I think it's a little bit an overlooked sort
 of security feature in general when it comes to Linux. And a
 lot of even experienced Linux administrators often haven't
 really heard of namespaces and how they can be used. Well,
 the next story, I guess, is sort of no good. Deed goes
 unpunished. Microsoft a couple of years ago came out with UI
 automation, user interface automation. That's an API that
 allows software running on a system to better and more
 simplistically interact with user interface elements. So,
 for example, software could read the content of Windows or
 interact with buttons and other user interface elements
 on the system. That, of course, makes it much easier
 to script GUI interactions. And, well, attackers are now
 taking advantage of this. They're now using malware in
 order to, for example, figure out what software is running
 on the system, then read the content of various Windows.
 And if they're, for example, seeing a browser window that
 is connected to an online banking website, well, they
 may take advantage of that and then steal credentials as the
 user enters them. Overall, this is not sort of a
 fundamentally groundbreaking new capability here that makes
 Windows a lot less secure. It just makes it a little bit
 easier for attackers. Attackers in the past have,
 for example, just used browser plugins in order to steal
 requests. That sort of works, too. It is a little more
 generic. It goes beyond browsers. And that just makes
 things a little bit easier for the attacker to interact these
 kind of interactions with the user interface. From a
 defensive point of view, Akamai recommends to basically
 monitor the UI automation core .dll. That's where you would
 see earliest requests and interactions like this
 happening. Basically, what software uses that dll would
 certainly be something worth paying attention to. And they
 have here a couple of queries in their blog that will allow
 you to select and query this interaction from Akamai's own
 tools. For your tool, whatever you're using to monitor your
 endpoints, you probably have to figure out how that
 translates. And then something a little bit different here.
 Usually, I don't really talk much about tools. But this
 tool that I just came across and find kind of useful. In
 particular, since we have so many vulnerabilities in APIs
 these days. This particular tool, AutoSwagger, is built
 for APIs that are using the OpenAPI or Swagger standard in
 order to publish documentation for their APIs. And then the
 tool does something that's actually really not all that
 terribly difficult. It reads the specification, which is
 meant to be machine readable, to then find particular
 vulnerabilities in the API. For example, various API
 endpoints that may be leaking PII without requiring
 authentication. Sounds like a tool that you probably should
 give a try if you are developing APIs or if you're
 testing them. The entire process is a little bit like,
 you know, what you have in SOAP with Whistles. OpenAPI or
 Swagger for REST APIs is often used to describe the APIs,
 again, in a standardized machine readable format. Well,
 and that's it for today. So thanks again for listening and
 talk to you again tomorrow. Bye.