What happened
Replaying the fake OAuth app intrusion, early 2025
- 1
Lure: a fake app's consent page
Early 2025 to June 2025Request-for-quote lures open the consent prompt of a fake Microsoft 365 app, which sends the target on to the phishing page whether they accept or cancel.
- 2
Capture: the Tycoon relay takes the session
Dates not statedA Tycoon page in the tenant's own branding relays the sign-in and captures the session, and the actor signs in to five accounts with axios user agents.
- 3
Persistence: an added authentication method
Date not statedIn the OneDrive-2025 case the actor adds an authentication method to the account, which Proofpoint reads as a likely persistence attempt.
More than two dozen fake applications were authorized by more than two dozen users in more than 20 tenants in early 2025, per Proofpoint. The Tycoon relay captured credentials, MFA-approved tokens and session cookies, and the actor took over five accounts, signing in to OfficeHome and account pages and adding an authentication method in one case; what it did with the accounts afterward is not reported.
Where Microsoft Entra controls fit
A Conditional Access policy that requires the Phishing-resistant MFA authentication strength for all users and all resources would have refused the relayed sign-in and the later sign-ins to the five accounts. Turning user consent off would have stopped the grants to the fake apps, but not the redirect to the phishing page, which ran whether the target accepted or canceled. A compliant device requirement, trusted named locations and limits on method registration would have capped the reach. The lures, sent from other organizations' real mailboxes and through SendGrid, sit outside every cloud control. Check whether your users can still consent to third-party apps, and whether your sign-ins still accept a phishable factor.
Could this have been prevented?
In hindsight, not as a whole. 2 controls would have refused a step outright, and 5 controls would have limited one. 1 step has no cloud-provider control. The counts below are control assessments, not attack stages: several can address the same step. The same replay with the controls in place before the incident shows where each phase would have been refused or bounded.
- 2
- Prevented
- 5
- Contained
- 1
- Not covered
How it could have been prevented
The same replay with the controls in place. Green marks where a step is refused, blue where it is bounded, and red where no control reaches it. Open a row to read the control, why it acts, and its caveat.
Replaying the fake OAuth app intrusion, early 2025
1 · Lure: a fake app's consent page
1 Contained1 Not covered2 · Capture: the Tycoon relay takes the session
2 Prevented2 Contained6 on other clouds3 · Persistence: an added authentication method
2 Contained3 on other cloudsKey facts
- Proofpoint first saw the activity in early 2025; the earliest fake app in its indicator table was first seen on January 27, 2025, and the activity was ongoing when Proofpoint published on July 31, 2025
- More than 50 impersonated applications appeared in the email campaigns, among them RingCentral, SharePoint, Adobe and DocuSign look-alikes
- In early 2025 more than two dozen fake applications were authorized by more than two dozen users in more than 20 tenants, for scopes such as basic profile, offline access, openid, email and profile; use of any grant for an API call was not reported
- Whether the target accepted or canceled the consent prompt, the flow went on through redirectors and a CAPTCHA to a Tycoon page that relayed the Microsoft sign-in and captured the credentials, the MFA-approved token and the session cookie
- Account takeover was confirmed in five cases, four through a fake Adobe app and one through an app named OneDrive-2025; the actor signed in to OfficeHome with axios user agents and added an authentication method in the OneDrive-2025 case
- Proofpoint reported the apps to Microsoft, which began changing the Microsoft 365 default to require admin consent for third-party apps in mid-July 2025, due to complete by August 2025
Sources
Every fact in this breakdown comes from the public reporting below. The replay is a model of that reporting, not a reconstruction of the environment it describes.
- 1.Microsoft OAuth App Impersonation Campaign Leads to MFA PhishingProofpoint Threat Research · Jul 31, 2025 · ResearchThe lures, the fake OAuth apps and their consent scopes, the same redirect on accept or cancel, the redirect chain, the Tycoon relay and what it captured, the five takeovers and their sign-ins, the added authentication method, the counts, the report to Microsoft and Microsoft's default change
The estate in the replay is illustrative and drawn at the level of attack stages; the victim's real environment is not public. Verdicts describe what a control does when it is in place before the incident. They are not an assessment of the victim's security program.