LiteLLM, the popular open-source gateway used to centralize API access for OpenAI, Anthropic, and other AI providers, has a critical pre-authentication SQL injection bug that attackers started exploiting just 36 hours after the security advisory went public. The flaw lets anyone who can reach the proxy port read all the API keys stored inside - including master keys, virtual keys, and provider credentials. The bug was in the bearer-token check: the token was concatenated into a SQL query instead of passed as a parameter. Sysdig saw the first attack at 04:24 UTC on April 26, hitting three tables that hold the most valuable secrets.
Researchers disclosed CVE-2026-3854, a critical GitHub Enterprise Server flaw that lets anyone with push access execute arbitrary commands on the GitHub server with a single git push. The bug is in how Enterprise Server handles repository hooks during push operations - a crafted commit message or filename bypasses the sanitization that normally prevents shell injection. GitHub patched it last week, but self-hosted instances need to apply the patch manually, and telemetry shows most haven't yet. Anyone with developer-level access to a vulnerable Enterprise Server can take over the entire instance, then pivot into every repository and CI/CD secret it hosts.
Microsoft confirmed yesterday that a Windows Shell spoofing flaw, CVE-2026-32202, is being exploited in the wild. The bug lets an attacker craft files that appear in File Explorer with fake names, icons, and paths - so a malicious .exe can show up looking like a benign PDF, leading users to double-click and run it. Microsoft patched the bug in the April 14 Patch Tuesday but only confirmed in-the-wild exploitation on April 28, raising urgency for any environment that hasn't deployed April patches. The flaw is particularly dangerous on shared file servers, USB drops, and email attachments - any path where users trust File Explorer to tell them what's what.
Microsoft quietly patched a privilege escalation flaw in Entra ID (formerly Azure AD) that let an attacker with a low-privileged service account take over any service principal in the same tenant - including high-value ones with admin consent grants. The bug was in how Entra ID validated role assignments during certain API calls: the validator checked whether the caller had any role on a service principal but didn't check whether that role authorized the specific action. Microsoft fixed the flaw on the back end, so customers don't need a patch - but the takeover scenario means anyone who exploited it before the fix could have created persistent backdoors via OAuth grants.
Researchers disclosed a critical unauthenticated remote code execution flaw in Hugging Face's LeRobot, the open-source framework used to train and deploy ML models on physical robots. CVE-2026-25874 sits in the framework's web interface, which by default listens on all network interfaces with no authentication - quick for demos, but a hard fail when the demo box ends up on a corporate network. There is no patch yet. Hugging Face has been notified but hasn't released a fix. Particularly serious because LeRobot is usually attached to actual robotic hardware, so a compromise can mean unsafe physical actions.
Researchers found a serious bug in VECT 2.0, a new ransomware family making the rounds: the encryption routine corrupts any file larger than about 131 KB instead of encrypting it reversibly. Files smaller than the threshold encrypt and decrypt normally; everything bigger gets permanently destroyed. Operators don't seem to know yet, so victims who pay get a working decryption tool that recovers small files and tells them the large ones are 'corrupted' - which they are, because VECT broke them on the way in. The bug affects Windows, Linux, and VMware ESXi variants. Any large file on a VECT 2.0-hit system is irrecoverable regardless of whether the ransom is paid.
Vimeo confirmed yesterday that user data was exposed when its analytics provider Anodot was breached. The video service hasn't said how many users are affected or what data was exposed beyond 'limited' account information, but Anodot's role suggests the leaked records include event-level user activity tied to Vimeo accounts: video views, account IDs, and the kind of telemetry analytics providers ingest. The pattern is the same as Citizens Bank, Frost Bank, Pitney Bowes, and now Vimeo: customer data leaks through a third-party vendor that the customer never directly signed up with.
cPanel disclosed a critical authentication bypass on Monday affecting every cPanel and WHM version - including end-of-life builds. CVSS 9.8. The bug let unauthenticated attackers log in as administrators by abusing how the cPanel session daemon writes session files during login. Hosting providers including Namecheap, KnownHost, hosting.com, HostPapa, and InMotion took cPanel and WHM offline globally for hours while patches deployed. Researchers at watchTowr published a working proof-of-concept on April 29. KnownHost reports possible targeted exploitation as early as February 23, 2026 - more than two months before disclosure.
North Korea's BlueNoroff group has built a self-reinforcing deepfake pipeline that turns each victim into the lure for the next attack. Arctic Wolf documented the pattern: attackers send a Calendly invite that looks like a normal business meeting, then quietly swap the Google Meet link for a typo-squatted Zoom URL. When the target joins, a fake Zoom interface secretly records their webcam feed while a clipboard-injection attack drops malware. The captured footage is mixed with AI-generated lookalikes (built using ChatGPT for synthetic portraits) and recycled into the next attack. Arctic Wolf found 950 files in BlueNoroff's media server. 80% of identified targets are crypto executives.
Pitney Bowes customer and employee data was leaked publicly after the company refused to pay ShinyHunters' extortion demand. Have I Been Pwned added the breach yesterday with 8.2 million unique email addresses, plus names, phone numbers, and physical addresses. A subset includes Pitney Bowes employee records with job titles - a useful starter pack for highly-targeted phishing against named staff. The data came from a misconfigured Salesforce Experience Cloud 'Guest User' permission that let unauthenticated visitors query CRM records directly. ShinyHunters had posted Pitney Bowes on its leak site April 18 with a three-day deadline.
TrustStrike Labs