A Fake Teams Booking Page Got an Employee to Approve the Attacker’s Microsoft Sign-In, Step by Step
Attackers posing as an executive at a partner firm used a fake Microsoft Teams scheduling page to trick a construction company employee into approving a device code sign-in, which handed over a valid session token. Here is how each step worked and the layers that would have stopped it.
Why Device Code Phishing Slips Past MFA, and the Layered Defenses That Stop It
Most phishing tries to steal a password. This attack never needed one. Instead it persuaded an employee at a construction company to type a short code into Microsoft’s genuine sign-in page, and in doing so approve a session for the attacker. Huntress’s security operations center caught the intrusion, but only after the attacker had obtained a valid session token. The walkthrough below is based on Help Net Security’s report on the Huntress findings.
First, what “device code” sign-in is
Device code sign-in exists for things like smart TVs and command-line tools that cannot easily show a full login screen. The device displays a short code; you enter it on a phone or PC at Microsoft’s real login page, complete MFA if asked, and the device is signed in. In device code phishing, the attacker starts that flow on their own machine and gets the victim to enter the code. The victim does everything on a legitimate Microsoft page, and the attacker receives the access tokens.
The attack, step by step
- A believable email. The message came from a typosquatted domain impersonating a real senior vice president at a different construction firm, offering an introductory call about a long-term supply partnership.
- A link that looked clean. The email linked to a Google share link, which then redirected to the lure. The attacker’s own domain never appeared in the email, so URL reputation checks had nothing to flag.
- A fake Teams scheduler. The landing page imitated a Teams meeting booking page and even showed the impersonated executive’s headshot.
- A code disguised as a booking reference. Clicking “Continue” quietly started a Microsoft device code request on the attacker’s server. The next page displayed the resulting code as a “Booking Reference,” with a countdown timer adding pressure.
- The real Microsoft page. Clicking “Copy” or confirm put the code on the clipboard and opened a pop-up to Microsoft’s genuine sign-in page. Entering the code there, and completing MFA if prompted, approved the attacker’s session.
- A tidy exit. As soon as the attacker’s backend saw a successful sign-in, the page closed the pop-up, showed “Verification complete” and sent the victim to a Calendly booking page, so nothing seemed amiss.
- Hiding from scanners. The phishing kit was delivered as AES-256-GCM-encrypted content with the decryption key sitting alongside it on the same page. Email scanners and crawlers that do not run JavaScript saw only an encrypted blob.
- Built-in retries. If Microsoft returned an error, which can happen when MFA is required, the page told the victim a new verification code was being prepared and started a fresh device code session.
- Blending in. The final login was routed through a residential proxy in New Orleans so it looked like ordinary consumer traffic.
Why no single control was enough
Each step defeated a different safeguard. The look-alike domain got past a quick glance, the redirect got past link reputation checks, the encryption got past scanners, the genuine Microsoft page got past “check the address bar” training, and the victim’s own MFA approval got past MFA. That is exactly why defense has to come in layers.
The layers that break the chain
- Layer 1: turn off what you do not use. If your business has no need for device code sign-in, block it with conditional access policies. If a few devices need it, restrict it to those accounts or locations and monitor it. In Microsoft 365, Entra ID conditional access is the usual place to do this.
- Layer 2: phishing-resistant MFA. Passkeys and hardware security keys are harder to talk someone into misusing than codes and push approvals. Prioritize admins, finance and executives.
- Layer 3: email defenses that look beyond the link. Look-alike domain detection, impersonation protection for executives and key partners, and sandboxing that follows redirects all help.
- Layer 4: a simple rule for people. A booking page, invoice or meeting invite should never ask you to type a code into a Microsoft login. Teach staff that unexpected code prompts are a reason to stop and call IT.
- Layer 5: watch sign-in logs. Huntress recommends reviewing unfamiliar device sign-ins and flags programmatic user agents such as python-requests or node in interactive sign-in logs as a sign of automated attacker infrastructure.
- Layer 6: a clean-up checklist. After a confirmed compromise, revoke sessions and check for rogue MFA methods, unknown registered devices, unauthorized app consents, suspicious mailbox rules and unusual outbound mail.
What this means for your company
Attacks like this target businesses that work with outside partners every day, where an introduction from a familiar company is routine. MFA is still essential, but approving a sign-in is not the same as knowing whose sign-in you approved. Close off sign-in methods you do not need, move key people to phishing-resistant MFA, and make sure someone is watching for unusual sign-ins at any hour. Each layer catches what the one before it misses.
Not sure whether device code sign-in is open in your Microsoft 365 tenant, or who would notice an odd login? Reach out to Layer 7 Data Solutions to discuss how we’ve helped companies similar to yours.
