The ongoing Shai-Hulud supply-chain campaign has struck again, this time trojanizing 19 Python packages on PyPI, many of them popular bioinformatics tools like Dynamo, Spateo, CoolBox, and Napari-UFISH that have been downloaded hundreds of thousands of times. Discovered by Socket, the wave pushed 37 malicious package versions from what looks like a single compromised maintainer, each carrying code that steals developer secrets such as cloud keys and tokens, then uses them to spread further. PyPI has quarantined affected releases. The credential-stealing behavior and tactics match earlier Shai-Hulud activity tied to the group TeamPCP, whose worm code leaked publicly last month.
Researchers at D3Lab warn that new versions of the NFCShare Android malware are spreading as fake updates for real banking apps, hosted on GitHub to look legitimate. Targeting customers of European banks, the malware shows a fake verification screen that tells victims to hold their payment card against the phone. It then uses the phone's NFC chip to read the card number, type, and expiry, and tricks the victim into typing their 4-digit PIN, sending it all to the attacker's server. That stolen data feeds NFC relay fraud, where criminals use it to make contactless payments or withdrawals. The malware only works if users sideload it.
Meta has confirmed that attackers took over 20,225 Instagram accounts by abusing a flaw in its AI-assisted account recovery tool, called High Touch Support. A bug meant the system never checked that the email address someone supplied actually belonged to the account, so an attacker could request a password reset for any account and have the link sent to their own inbox, then walk in, unless the target had two-factor authentication on. High-profile accounts, reportedly including the Obama White House and US Space Force personnel, were hijacked and sold on the dark web. Meta has secured the accounts and is fixing the verification check before relaunching the tool.
Meta says it caught and shut down fresh spear-phishing attempts linked to Israeli spyware maker NSO Group that tried to lure WhatsApp users into clicking malicious links leading to sites outside the app, mirroring the one-click attacks NSO has used to plant its Pegasus spyware. Meta also found and removed NSO-created test accounts and groups, and published the malicious domains involved. The company is now asking a US federal court to hold NSO in contempt for violating the permanent injunction issued last year barring it from targeting WhatsApp. High-risk users such as journalists, activists, and officials are the usual targets of this kind of mercenary spyware.
Fortinet has uncovered a new botnet called C0XMO, built from the long-running Gafgyt malware family, that breaks into devices by exploiting an old flaw (CVE-2021-27137) in the UPnP service of DD-WRT router firmware. A booby-trapped network request gives the attacker code execution with no login needed. Once in, C0XMO digs in with hidden files and cron jobs that re-run it every 15 minutes, then hunts down and deletes rival botnets and even researchers' security tools to keep the device to itself. A separate scanner spreads it across many chip types (ARM, MIPS, x86, and more), and infected devices are wired up to launch 19 kinds of denial-of-service floods.
Mandiant has detailed how the extortion crew Silent Ransom Group (also tracked as Luna Moth and UNC3753) is breaking into US law firms and other professional-services companies through phone calls rather than malware. Attackers send a harmless-looking invoice or data-migration email, then call the target pretending to be internal IT support, talk them into starting a screen-share, and get them to install a remote management tool that hands over access. From there, Mandiant has seen data located, staged, and stolen in under an hour. The group skips encryption entirely, instead threatening to leak stolen files unless paid. A recent FBI alert added in-person office visits to the playbook.
The Five Eyes intelligence agencies (US, UK, Canada, Australia, and New Zealand) issued a joint bulletin, "Safeguarding Our Secrets," warning that Chinese military intelligence officers are posing as recruiters on sites like LinkedIn, Indeed, and Upwork. Fronting as think tanks, consultancies, or HR firms, they post fake jobs such as foreign-policy or defense-analyst roles, then use the interview process to pressure targets into handing over classified or non-public information. The agencies say current and former government, military, defense-contractor, research, and journalist personnel are all in scope, with extra focus on those tied to the Indo-Pacific. The goal is harvesting privileged military, political, and economic intelligence.
RCI Hospitality, one of the largest US adult-nightclub operators, has confirmed that a breach exposed the personal data of 40,178 people, mostly independent contractors. Attackers got in through an insecure direct object reference (IDOR) flaw on one of the company's IIS web servers, a common web bug where simply changing an ID number in a web address lets you pull up someone else's record. The intrusion began March 19 and was spotted four days later. Stolen data includes names, dates of birth, Social Security numbers, and driver's license numbers. RCI says no customer or financial systems were touched, and the data has not yet appeared publicly.
A security startup's autonomous AI agent scanned FFmpeg, the open-source media library built into countless video and audio tools, and turned up 21 previously unknown bugs, each with working proof-of-concept code that crashes or corrupts memory when the software processes a malicious media file. Several flaws are 15 to 20 years old; one dates back to 2003. Nine already carry CVE numbers (CVE-2026-39210 through CVE-2026-39218), and the rest are fixed but not yet numbered. The whole run cost about $1,000. Because FFmpeg sits inside browsers, media servers, and apps everywhere, any product that decodes untrusted video could be at risk.
Google shipped Chrome 149 with fixes for 429 security bugs, the most ever in a single Chrome release. More than 100 are rated critical or high. The worst, an out-of-bounds read and write in the ANGLE graphics engine that Chrome uses to render web pages, lets a booby-trapped website break out of the browser's protective sandbox and run code on the victim's computer; Google paid a $97,000 bounty for it. None are confirmed under attack yet, but a sandbox escape is the kind of bug attackers race to weaponize, so patching before that happens matters.
TrustStrike Labs