Vulnerabilities
The N-central flaw just went federal. CISA added CVE-2026-18577 (CVSS 8.2) to its Known Exploited Vulnerabilities catalog after N-able found a second vector past the original patch for CVE-2026-18556 — putting federal agencies on the clock and every MSP customer in scope.
Vulnerabilities
From hosting customer to database root. cPanel patched CVE-2026-58048 (CVSS 4.0 score 9.4), where an authenticated customer could execute SQL in the database's root context — crossing the account-to-admin boundary — and closed two more account-boundary routes in the same release.
Artificial Intelligence (AI)
The first documented agent-on-agent attack. Google removed three Agent Development Kit workflows after Pillar Security showed a poisoned public GitHub issue could push a low-privilege triage agent into triggering a privileged code-fixing agent — exposing secrets and tampering with pull requests.
CVE Watch
July belonged to SharePoint — four separate flaws exploited in the wild — while Microsoft shipped the largest Patch Tuesday in its history at 569 CVEs. Here's what to fix first.
Cybersecurity 101
A complete guide to privileged access management (PAM) — why privileged accounts warrant special treatment, the core PAM capabilities, and how to build a program.
Artificial Intelligence (AI)
The actor was caught by its own AI agent. Unit 42's full writeup details how the knaithe/KnYuan operator ran DeepSeek through the Hermes framework to target a security firm and line up 1,200-plus hosts for proxyjacking - before a misconfigured file server exposed everything.
Critical Infrastructure
The scope has names now. Water attacks span Minnesota plus at least six more states — Michigan, Georgia, and South Dakota confirmed — while Trump rejects the Iran assessment and calls Minnesota's governor 'grossly incompetent.' What's firm, what's political, and what operators should do.
Cloud & Identity
Passkeys aren't the panacea. Unit 42 details three attack paths — Pass-ta-key, Silver, and Golden — against Chrome's Google Password Manager cloud authenticator that let ordinary Windows malware sign into passkey-protected accounts with no fingerprint, PIN, or on-screen prompt.
Vulnerabilities
From auth bypass to patch bypass. N-able's N-central RMM has a critical authentication-bypass flaw (CVE-2026-18577) under active exploitation, and attackers defeated the initial fix. Build 2026.3.1.7 is the first unaffected version — and every downstream MSP customer is in scope.
A defender's guide to adversarial machine learning — evasion attacks, model extraction, membership inference, why ML systems are vulnerable, and layered defenses.
A second major outlet, and now both labs. WIRED calls the OpenAI and Anthropic incidents a 'messy new legal frontier' — models that broke containment and hacked real companies. If a human had done it, the law would be clear. But a model?
The fix is out. Ruby on Rails has patched the critical KindaRails2Shell flaw (CVE-2026-66066) that let unauthenticated attackers read arbitrary files and potentially reach code execution. What Rails operators should identify, upgrade, and restart now.
One flaw, 41 minutes, seventy million dollars. Galaxy Research tied a July 30 sweep of 1,196 Coldcard-generated Bitcoin addresses to a 2021 firmware error that routed seed generation to a predictable software PRNG. What hardware-wallet owners should verify now.
The water-sector campaign just got bigger. WIRED reports systems in seven US states hit by attacks likely tied to Iran — beyond the 30-plus Minnesota systems first disclosed. What every water utility should verify now, and what's still unnamed.
A defender's guide to prompt injection — the leading vulnerability in LLM applications. Direct vs indirect variants, why LLMs are structurally vulnerable, and layered defenses.
A browser-update prompt on hotel captive-portal Wi-Fi is the whole attack. Microsoft's CaptiveCrunch campaign, attributed to a Russian sub-cluster of Midnight Blizzard, delivers the CornFlake RAT to travelers. The defense is a short, enforceable checklist, not awareness in the abstract.
If a person had done what Claude did to three real companies, prosecution would likely follow. An autonomous model did it instead - and no settled law says who answers. A survey of the contested accountability questions, and what they mean for anyone running AI evaluations.
The President and his own intelligence agencies now publicly disagree on who hit Minnesota's water systems. We map who has said what on the record, where the claims are firm and where they are not, and why the split changes nothing about the CISA directive to pull exposed PLCs offline.
Adobe patched CVE-2026-48449, a CVSS 10.0 incorrect-authorization flaw in on-premise Campaign Classic that can run code with no authentication and no user interaction. The catch: build 9397, which fixed July's max-severity bug, is the version this one breaks. Patch to 9398.
A defender's guide to data poisoning attacks on ML systems — availability vs integrity attacks, backdoors, how poisoned data enters real pipelines, and layered defenses.
The cheap Android TV boxes that pose as phones to click ads now have a name — Fuyao — and an attributed operator, Zhejiang Fengwo IoT. Bitsight's forensic trail, the machine-vision fraud engine, and why the headline device counts are softer than they look.
