A development flag left switched on in production builds of several Microsoft 365 Android apps disabled the check that limits account-token sharing to trusted Microsoft apps.
Any other app on the same phone could ask for the signed-in user’s token and get it, then read email, open files, browse the calendar, and send messages as that user. No password, no login screen, no permission prompt.
Protect your privacy by Mullvad VPN. Mullvad VPN is one of the famous brands in the security and privacy world. With Mullvad VPN you will not even be asked for your email address. No log policy, no data from you will be saved. Get your license key now from the official distributor of Mullvad with discount: SerialCart® (Limited Offer).
➤ Get Mullvad VPN with 12% Discount
Microsoft has patched it, and if you run Microsoft 365 apps on Android, update them.
The bug, which Enclave calls FlagLeft, hit Word, PowerPoint, Excel, Microsoft 365 Copilot, Microsoft Loop, and OneNote, six apps with billions of downloads between them. Teams shipped with the same flag set to false and were not affected, which Enclave reads as a slip rather than a design.
Microsoft 365 apps share account access on purpose, so signing into Word means you do not sign in again for PowerPoint. The handoff is supposed to verify who is asking and turn away anything that is not a trusted Microsoft app.
Enclave’s Yanir Tsarimi and Ofek Levin found the check was being skipped because of a single line left in the shipping code: setIsDebugMode(true). The flaw sat in a shared Microsoft SDK, so the same hole showed up in app after app.
The tokens handed over were FOCI tokens, the family refreshes tokens Microsoft uses for single sign-on across its apps. They can be refreshed and reused over long stretches, and the resulting traffic looks routine in logs. From the user’s side, nothing visible happens.
Enclave built a working proof of concept that pulled tokens through an unverified third-party app and read email with them. Microsoft classifies these as local spoofing flaws; in plain terms, a malicious app already on the device is all it takes.
Microsoft issued four CVEs on May 12, all classed as spoofing under improper access control (CWE-284): CVE-2026-41100 for Microsoft 365 Copilot (CVSS 4.4), CVE-2026-41101 for Word (CVSS 7.1), CVE-2026-41102 for PowerPoint (CVSS 7.1), and CVE-2026-42832 for Excel (CVSS 7.7). The four CVEs cover Copilot, Word, PowerPoint, and Excel.
Enclave reported the same flaw in Loop and OneNote, but neither got a separate CVE in the May batch. NVD lists the patched Word build for Android as 16.0.19822.20190, with earlier versions affected. The other apps were fixed through the same Google Play updates.
Nothing in Microsoft’s May Patch Tuesday release was listed as publicly known or exploited, and there is no public evidence that the flaw was used before the fix.
What to do? Update Word, PowerPoint, Excel, Microsoft 365 Copilot, Loop, and OneNote from Google Play. Security teams managing Android fleets should push the updates through MDM and confirm devices are off builds earlier than 16.0.19822.20190.
The patch closes the hole, but it does not retroactively kill tokens that an attacker may already hold. FOCI refresh tokens outlive an app update, so for accounts on devices that ran an old build alongside untrusted apps, it is worth revoking refresh tokens and forcing a fresh sign-in.
Found this article interesting? Follow us on Google News, Twitter and LinkedIn to read more exclusive content we post.
Some parts of this article are sourced from:
thehackernews.com
Autonomous AI Tool Finds 2-Year-Old RCE Flaw in Redis (CVE-2026-23479)
Jun 03, 2026
Vulnerability / Cloud Security
Redis has patched a use-after-free in its blocking-client code that lets an authenticated user run arbitrary OS commands on the machine hosting the database. The flaw was found by an autonomous AI tool built to hunt bugs in large codebases. Tracked as CVE-2026-23479 , the flaw was introduced in Redis 7.2.0 and remained in every stable branch until the May 5 fixes, unnoticed for over two years. NVD rates it 8.8 under CVSS 3.1; Redis lists it as 7.7 under CVSS 4.0. It was reported by Team Xint Code, and a complete technical write-up is now public. The cloud footprint makes this worse. Wiz's analysis, published with the exploit writeup, puts Redis in a large majority of cloud environments, with most of those instances running without a password. The exploit needs an authenticated session, but in a default deployment, the default user already holds every privilege the chain requires. The flaw lives in unblockClientOnKey() in src/blocked.c , which fires when a key event wakes …