Daily brief: Saturday 31 October 2026
TeamViewer patches a flaw that lets a malicious host attack the person connecting to it, a trojanised Claude Code extension hit Cursor users, CISA confirms GitLab attacks, and WordPress and Linux need updates.
TeamViewer 15.83 fixes a flaw that lets a remote computer attack the person controlling it
TeamViewer's bulletin TV-2026-1011 covers three high-severity flaws, rated up to CVSS 8.8. The one getting attention, CVE-2026-111585, lets a malicious or compromised host crash the viewer and, depending on the platform, run code with the viewer's privileges. It affects the Full Client (including Portable) before 15.83 on Windows, Linux and macOS, and the Android Remote Control app. TeamViewer says it is not aware of any exploitation, and there is no workaround, only the update.
Source: TeamViewer security bulletin TV-2026-1011 · The Hacker News
Fake Claude Code update on the Cursor marketplace planted an SSH backdoor
A malicious version 2.1.313 of the Claude Code extension was published to the Cursor Marketplace under Anthropic's real extension ID on 29 October. It added an attacker's key to the user's SSH authorized_keys file, tried to enable SSH and open a reverse tunnel, set up persistence and sent basic host details to the attacker. Cursor says at most 5,000 users installed it during a roughly 11-hour window, and that the version was not made by Anthropic. There is no known evidence it reached Open VSX or Microsoft's VS Code Marketplace; Anthropic has released 2.1.314.
Source: The Hacker News
CISA confirms attacks on the critical GitLab flaw
CISA added CVE-2026-97229, a CVSS 10.0 GitLab flaw that can give an unauthenticated attacker code execution through a crafted regex in a CI configuration, to its Known Exploited Vulnerabilities list on 29 October. It also added the Cisco ASA and FTD denial-of-service flaw covered yesterday. Alpine Linux has described an attempt on its own GitLab server: it found no proof the attack worked but rebuilt the server anyway. The bug is in the Ruby regex engine, so self-compiled installs also need Ruby 3.3.13, 3.4.12, 4.0.9 or later.
Source: The Hacker News · The Hacker News: Alpine Linux GitLab incident
WordPress 7.1.4 fixes a core flaw that is already drawing tens of thousands of attacks
WordPress 7.1.4, released on 28 October, fixes a high-severity (CVSS 8.1) flaw in WordPress core affecting versions 4.7 to 7.1.3. During a full core upgrade, files are unpacked to a predictable folder under wp-content/upgrade, and an unauthenticated attacker who hits that window can reach PHP code execution. Wordfence says it blocked 62,969 attacks aimed at the flaw in 24 hours. No CVE had been assigned when the report was written.
Source: The Hacker News
Linux kernel TCP flaw: some distributions have fixes, others are still working on them
CVE-2026-98432 is a race condition in IPv4 TCP connection setup that dates back to Linux 4.4. The kernel CVE team rates it 9.8, while Red Hat, SUSE, Oracle and Amazon give lower scores. Anthropic says Claude wrote a working exploit that gives a remote attacker root on servers with a multipath TCP listener, but no attacks have been reported. Debian, Oracle Linux and some SUSE and Amazon Linux kernels are fixed; Ubuntu and RHEL 8 to 10 fixes were still pending on 29 to 30 October.
Source: The Hacker News
Fake ChatGPT and Claude desktop apps on GitHub install a crypto miner
Check Point Research found a campaign it calls Yoba that uses fake GitHub projects offering desktop apps for ChatGPT and Claude. The download holds a real, signed OpenAI Codex or Claude executable next to a malicious winmm.dll, which the signed program loads. The chain sets up a scheduled task and ends by running the XMRig cryptocurrency miner. It waits about 12.7 days in total before acting, to get past sandboxes.
Source: The Hacker News
Google announces Gemini 4 Carbon, aimed first at cyber defenders
Google unveiled Gemini 4 Carbon on 29 October, describing it as a frontier model with strong security, coding and agent abilities. It is first being released to a limited group of cyber defenders, then to trusted partners through Google's Fairwind Program. Google reports scores of 88.2% on SEC-bench Pro for finding vulnerabilities and 71.0% on CWE-bench for patching. These are Google's own figures.
Source: The Hacker News
Weather
Live forecastLoading the forecast…