Maximum-Severity Exploit Hits Corporate Networks

Arista Networks makes networking equipment and software used by enterprises to manage wide-area networks across multiple office locations. Their VeloCloud Orchestrator (VCO) is the central control platform that manages all of those connected sites from one place.

That platform has a CVSS 10.0 flaw being actively exploited right now, according to The Hacker News. The vulnerability is an OS command injection flaw — meaning an attacker can send specially crafted requests to the VCO web interface and have the server execute system-level commands as if they typed them directly into the machine. No login required.

Arista confirmed the flaw was discovered externally and is already being weaponised. The company shared three IP addresses actively involved in the attacks: 8.19.75.217, 206.72.242.124, and 206.72.242.162. Block these immediately if you manage a VCO instance.

What makes this particularly serious: a compromised orchestrator can give attackers a path into every VeloCloud Edge device it manages — potentially every branch office connected to the network.

Affected on-premises versions include VCO 5.2.x, 6.1.x, 6.4.x, and 7.0.x before their respective fixed releases. Cloud-hosted versions have already been patched.

CISA added this to its Known Exploited Vulnerabilities catalog, requiring federal civilian agencies to apply the patch by 30 July 2026.

What to do: If you manage an on-premises VCO instance, patch to the fixed release immediately. If patching is not immediately possible, restrict web interface access to trusted administrative networks only and check your logs against those three attacker IP addresses.

A researcher at STAR Labs has published a working exploit that lets any local user become root on CentOS Stream 9, and they used AI to help build it, according to The Hacker News.

The flaw, CVE-2026-53264, is a use-after-free race condition — a bug where two processes compete to access the same memory, and one reads it after the other has already freed it — sitting inside the Linux kernel's network traffic-control subsystem.

The AI assisted with finding the bug, generating a proof of concept, and widening the timing window needed to trigger the flaw reliably. Researcher Lee Jia Jie was careful to note that human judgement was essential throughout.

This is local exploitation only — an attacker needs existing access to the machine first. The exploit also requires specific kernel configuration options, narrowing real-world exposure. But the full source code is now public, which raises the urgency considerably.

What to do: Linux administrators should update to a fixed kernel release (5.10.259, 5.15.210, 6.1.176, 6.6.143, 6.12.94, 6.18.36, or 7.0.13) and install the version provided by their distribution rather than compiling upstream.

What TeamCity is: TeamCity is a continuous integration and delivery server made by JetBrains — software used by development teams to automatically build, test, and deploy code.

What it is: A critical authentication bypass that allows an unauthenticated attacker to reach internal server functionality and execute operating system commands — all without ever logging in.

Who's at risk: Any organisation running TeamCity On-Premises with an internet-facing or network-accessible server. TeamCity Cloud instances are already patched.

CVSS: 9.8 — Critical. Patch today.

Root cause: The flaw sits in the agent polling protocol, a communication channel normally used by build agents (automated workers) to check in with the TeamCity server. The server failed to properly enforce authentication checks on this channel, meaning an attacker can send requests through it as if they were a legitimate agent and reach privileged functionality they should never be able to access. Think of it like a staff entrance with a broken lock — the front desk checks every visitor, but this side door waves anyone through.

Attack vector: An attacker with HTTP or HTTPS access to the TeamCity server sends crafted requests through the agent polling endpoint, bypassing authentication entirely. Once past that check, they can issue operating system commands that run with the same permissions as the TeamCity server process itself — which in many deployments means access to stored credentials, build configurations, source code, and deployment pipelines. No user interaction and no existing account is required.

Detection strategies:

1. Review TeamCity server logs for unexpected or unauthenticated requests to agent polling endpoints, particularly from unfamiliar IP addresses.
2. Check for unexpected outbound connections from the TeamCity host, especially to unusual destinations.
3. Audit recently modified build configurations, stored credentials, or administrator accounts for changes you did not make.

Recommended actions:

1. Upgrade to TeamCity On-Premises version 2025.11.7 or 2026.1.3 immediately.
2. If upgrading now is not possible, install JetBrains' security patch plugin (available for versions 2017.1 and above) as an interim measure.
3. Place the TeamCity server behind a VPN or restrict network access so it is not reachable from the public internet.

A supply chain attack (where hackers poison a software package that developers trust and use, sending malicious code downstream to every project that installs it) is one of the most effective ways to compromise thousands of organisations at once.

GitHub and PyPI — the Python Package Index, the main repository where Python software packages are published — have both introduced new protections, according to Security Week.

GitHub's automated dependency tool, Dependabot, now waits three days before updating your project to a brand-new release. That gap gives security researchers time to catch a poisoned version before it reaches your codebase. PyPI is now blocking new files from being added to releases older than 14 days, closing a gap where an attacker with a stolen publishing token could quietly poison a long-trusted package version.

What to do: If your team uses Dependabot, check that your dependabot.yml configuration is not overriding the new cooldown behaviour. If you maintain Python packages on PyPI, no action is needed — the protection applies automatically.

Senator Ron Wyden has written to CISA, the Office of Management and Budget, and NIST demanding the removal of all legacy public-facing VPNs from federal networks, citing repeated exploitation by Russian and Chinese state-backed attackers through products from Cisco, Fortinet, Ivanti, and Check Point, according to The Record.

The proposal calls for a two-year deadline to replace ageing remote-access systems with zero-trust architecture — a security model that continuously verifies every user rather than trusting anyone already inside the network. For small businesses, the signal is clear: if legacy VPNs are too risky for federal agencies, they are worth reconsidering for your organisation too.