Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

Plaintext with Rich

Plaintext with Rich artwork

Device-Code Phishing: The Real Login Page Trap

Oct 16, 20269 minAudio
ShareXLinkedInFacebookWhatsAppEmail

Device-code phishing starts with an uncomfortable possibility: the sign-in address you checked really does belong to Microsoft. The document you're expecting comes with a code, the page looks right, and the question that matters is still hiding in the request. Plaintext with Rich follows the approval from the person entering that code to the software waiting on the other end. A television sign-in and a parcel collection slip help explain device-code flow without turning your commute into a technical certification. The episode draws on the Federal Bureau of Investigation's May 21, 2026, Kali365 advisory to unpack how criminals used genuine Microsoft pages in requests for Microsoft 365 access. You'll hear what access tokens and refresh tokens do, why the code you enter connects to a separate pending request, and how multi-factor authentication fits into the picture. There's also a leadership question about whether awareness training shows employees enough legitimate setup activity to recognize a request that doesn't belong. If you handle shared documents, help colleagues with account access, or write security guidance, this is for you. No background in identity systems required, just some curiosity about what a successful sign-in actually approves. Listen for the missing context behind that familiar page. One Topic, Ten minutes, No panic. Is there a topic/term you want me to discuss next? Text me!! YouTube more your speed? → https://links.sith2.com/YouTube Apple Podcasts your usual stop? → https://links.sith2.com/Apple Neither of those? Spotify’s over here → https://links.sith2.com/Spotify Prefer reading quietly at your own pace? → https://links.sith2.com/Blog Join us in The Cyber Sanctuary (no robes required) → https://links.sith2.com/Discord Follow the human behind the microphone → https://links.sith2.com/linkedin Need another way to reach me? That’s here → https://linktr.ee/rich.greene

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Transcript

From the show. Tap a time to play from there.

SPEAKER_00You're waiting for a document before a meeting, and an email arrives with a code to view it. You check the sign and address carefully because you've heard that advice a hundred times, and it really does belong to Microsoft. No creative spelling, no mystery website, and the meeting hasn't even started yet. You're feeling almost annoyingly responsible, which makes this an awkward moment to discover that the address wasn't the only thing worth checking. Welcome to plain text with the rich.

SPEAKER_00Let's look at how a real sign-in page can help someone else get into your account and the extra question that changes the decision. In plain text, device code phishing tricks you into approving access for a device or application someone else controls. The criminal supplies a code, you complete a genuine sign-in process, and the access goes to the wrong place. The deception is in what you've been asked to do. I find this one particularly irritating. We ask people all the time to check the address, they check the address, and then we discover our advice needed another sentence.

SPEAKER_00So let's give you that sentence, along with enough of the explanation to make it stick. First, there's a perfectly useful feature underneath all of this. Think of signing into a television app using your phone. The television shows a short code, you visit the service's website on your phone, and you approve connecting television to your account. Anyone who has tried entering a long password with a television remote understands why somebody invented this particular arrangement. That process is called a device code flow, meaning the sequence that lets one device get access after you sign in on another.

SPEAKER_00In the legitimate version, you started the setup and you know which device is waiting. Those two facts are doing a lot of the work. Now imagine a parcel counter that lets you approve someone else collecting your delivery. A collection slip identifies the pickup arrangement, and the clerk checks your identity before accepting your approval. The building is real, the clerk is real, and your identification is valid. But suppose someone handed you their collection slip while telling you it was a form to check your delivery time.

SPEAKER_00The clerk can verify who you are without knowing that somebody lied to you about the form. You've authorized a collection you didn't mean to authorize. The comparison only goes so far. A sign-in page isn't literally handing over a package, and you haven't necessarily disclosed your password to anyone. The useful part is the relationship. Identifying the person who approves something doesn't establish that person understands the request. Now, on May 21st, 2026, the FBI warned about Cali 365, a phishing service criminals could subscribe to.

SPEAKER_00Its advisory described emails impersonating familiar document sharing and productivity services, supplying a device code and directing people to a legitimate Microsoft page. Completing that request could authorize the attacker's access to the person's Microsoft 365 account, including things such as Outlook, Teams, as well as OneDrive. Here's what's happening underneath. An attacker starts a device code process with software they control. The service supplies codes for that pending request, and the attacker sends you the short one you're meant to enter.

SPEAKER_00You sign in and approve the request, believing you're doing what the email promised. The waiting software then receives an access token, a digital proof that lets it make permit a request to the service. Microsoft hasn't accidentally sent your password to the attacker. The approved process has delivered access to the software that started it. There's a small detail here that I think makes the whole thing click. The short code you type is typically or technically the user code.

SPEAKER_00Behind the scenes, there's a separate device code that the waiting software uses to check whether you finished. You don't need to memorize either term, just understand that the code in front of you connects your approval to a request already waiting somewhere else. That somewhere else is the part the email hopes you won't think about. Now, you may also hear about a refresh token, which can let software obtain further access tokens without another full sign-in.

SPEAKER_00The FBI described attackers obtaining both kinds. That helps explain why access can continue beyond the moment you close the browser. Although the access available and its duration depend on permissions as well as security policies. Now, if you're wondering about multi-factor authentication, or you know, MFA, that's checking identity using more than one kind of evidence, such as a password and maybe an authenticator of some sort, keep using it. In this trick, you can pass the identity check yourself while approving an attacker initiated request.

SPEAKER_00Strong evidence that you are you doesn't establish why you're authorizing something. And again, episode 11 had the broader phishing discussion. The extra lesson here is to check the purpose of a sign-in as carefully as its destination. A genuine address answers which service you're visiting, right? You still need to know what you're asking that service to allow. For leaders, that means awareness material needs examples of legitimate requests too.

SPEAKER_00Show people what approved device setup looks like in your organization, who initiates it and where to ask when the screens don't match the task. Telling an employee to recognize the wrong process is much more reasonable after you've shown them the right one. And give that question a welcome reception. If somebody says, why am I connecting a device when I'm trying to read a document? That's useful judgment. I'd much rather hear that during a busy meeting than discover we've trained everyone to hurry through anything, again, just with a familiar logo.

SPEAKER_00So let's go back to that meeting. The document is supposedly waiting, the code is sitting in your email, and you've reached a perfectly genuine sign-in page. Before entering anything, again, try asking, what's waiting for me on the other end of this? With your television, you can answer that. You started this setup and the television is waiting for you to finish. Possibly while someone asks you whether you've tried pressing the button harder. With this unexpected document email, the answer is missing. You haven't started connecting a device, so why are you being asked to finish connecting one?

SPEAKER_00Read what the page says you're approving. If it's a device or application you didn't intend to connect, leave the request unfinished. You can open your usual workplace support portal or call a known support number and ask, is this device setup actually required to get this document? That's a question someone can look into. You know, is Microsoft a real company? Of course. Gives them considerably less to work with. All right. An organization can also reduce how often people face that decision.

SPEAKER_00Ask your identity administrator where device code flow is actually needed. Microsoft's guidance recommends blocking it wherever possible, with carefully limited exceptions, after auditing legitimate use and testing the policy. Approved work still needs to function. The point is to understand which processes justify keeping this option available. Now, if you've already approved an unexpected request, again, contact your account provider or workplace security support promptly and say you've entered a device code and completed the approval.

SPEAKER_00That detail helps them investigate the access involved. Provider screens and protections will change, so use the provider's current guidance for the exact controls. The habit survives those changes. Know what started the request and what your approval will connect. A real page can carry out a request you didn't intend. Check the service, read the action, and account for the device waiting on the other end. As always, send me your questions by email, DM, or comment. Deliver a note to the world's most suspicious parcel counter addressed to who actually requested this.

SPEAKER_00Every message gets read and answered. Please leave actual sign-in codes out of the correspondence. If you're enjoying yourself, go ahead and hit subscribe or follow or whatever your listening platform utilizes. It's the single best way to make sure that you don't miss our next episode. If this helped, hey, share it with someone who'd benefit. As always, this has been Plain Text with Rich. One topic, 10 minutes or less, no panic. I'll see you next time.

Open the transcript file

Hosts and Guests

Nobody is credited on this episode yet.

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. After the Click: How to Report a Security Mistake

    Oct 30, 20269 minAudio

    A security mistake can leave you writing and deleting the same message to your manager. You suspect something went wrong, but finding the words feels harder than it should. Plaintext with Rich explores the first useful...

    Credit the people on this episode

  2. Task Scams: Why You Shouldn't Pay to Get Paid

    Oct 23, 20269 minAudio

    Task scams can make an evening of online assignments feel like money earned. Then the withdrawal request turns into another demand, and the balance you've been watching starts to mean something different. Rich traces...

    Credit the people on this episode

  3. Browser Extensions: What Did You Give Permission To?

    Oct 9, 20268 minAudio

    Browser extensions can stay in your working day long after you've forgotten why you installed them. That tiny toolbar icon may represent a bigger arrangement than you remember making. Rich explores what extension...

    Credit the people on this episode

  4. ClickFix: When a CAPTCHA Asks You to Run a Command

    Oct 2, 20269 minAudio

    You came for a conference agenda, and the website wants proof you're human. ClickFix can hide behind that familiar check, with instructions that deserve a much closer look. Rich follows the moment a fake CAPTCHA, a test...

    Credit the people on this episode

  5. Last Seen - When the Phones Woke Up (Episode 4)

    Sep 25, 20265 minAudio

    Episode 4: When the Phones Woke Up Finding one another doesn’t fix the city. There are still people waiting for news, families searching for loved ones, and neighborhoods trying to function without the systems they...

    Credit the people on this episode