12/31/2026-01/07/2026 ISE Security Vulnerability, Ni8mare Flaw Gives Unauthenticated Control Of n8n Instances, New Veeam Vulnerabilities And More

1. Cisco Patches ISE Security Vulnerability After Public PoC Exploit Release

Cisco has issued patches for a medium-severity flaw, CVE-2026-20029 (CVSS: 4.9), in its Identity Services Engine (ISE) and ISE Passive Identity Connector. This vulnerability, present in the licensing feature, allows an authenticated administrator to read arbitrary files via malicious XML uploads. A public proof-of-concept exploit is available. Affected versions include releases earlier than 3.2 and specific 3.2 to 3.4 releases; version 3.5 is not vulnerable. No workarounds exist. Concurrently, Cisco fixed two other medium-severity Snort 3 bugs—CVE-2026-20026 (denial-of-service) and CVE-2026-20027 (information disclosure)—affecting multiple products. Given frequent targeting of Cisco vulnerabilities, users must apply updates promptly for protection.

2. Ni8mare Flaw Gives Unauthenticated Control Of n8n Instances

A critical vulnerability in the n8n workflow automation platform, tracked as CVE-2026-21858 (CVSS 10.0) and dubbed Ni8mare, allows unauthenticated attackers to fully compromise affected instances. Discovered by Cyera researchers, the flaw enables arbitrary file read by abusing how n8n Webhooks handle uploaded data in certain form-based workflows.

The issue arises when workflows process files without validating the request’s Content-Type. Attackers can craft non-multipart requests and manually define file paths, tricking n8n into copying and exposing sensitive system files such as /etc/passwd. With access to configuration files and the local SQLite database, attackers can extract authentication secrets, forge an admin session cookie, and bypass login protections.

Once authenticated as an admin, attackers can achieve full remote code execution using built-in workflow nodes. The vulnerability affects all n8n versions up to 1.65.0 and was fixed in version 1.121.0 (November 2025). A compromised n8n instance can expose credentials, tokens, and connected systems, making the impact severe.

3. New Veeam Vulnerabilities Expose Backup Servers to RCE Attacks

Veeam has released security updates to fix multiple flaws in its Backup & Replication (VBR) software, including a high-severity remote code execution vulnerability tracked as CVE-2025-59470. The flaw affects Veeam Backup & Replication version 13.0.1.180 and all earlier v13 builds.

The vulnerability allows attackers with Backup or Tape Operator roles to achieve remote code execution as the postgres user by sending malicious parameters. While initially rated critical, Veeam downgraded the issue to high severity because exploitation requires highly privileged access. Two additional flaws were also fixed: CVE-2025-55125 (high) and CVE-2025-59468 (medium), both enabling RCE under specific conditions.

The issues were patched in Veeam Backup & Replication 13.0.1.1071, released on January 6. VBR is widely used by enterprises and managed service providers and is frequently targeted by ransomware groups, as compromising backup servers can enable data theft and prevent recovery.

4. Critical AdonisJS Bodyparser Flaw (CVSS 9.2) Enables Arbitrary File Write on Servers

Users of the @adonisjs/bodyparser npm package are urged to update after disclosure of a critical path traversal vulnerability that could allow arbitrary file writes on servers. Tracked as CVE-2026-21440 (CVSS 9.2), the flaw affects AdonisJS multipart file uploads when developers use MultipartFile.move() without sanitizing filenames or providing the options parameter.

In such cases, attackers can supply crafted filenames containing traversal sequences, enabling them to write files outside the intended upload directory and potentially overwrite sensitive files. If application code or configuration files are overwritten and later executed, remote code execution may be possible, depending on deployment and permissions. The issue affects versions ≤10.1.1 and ≤11.0.0-next.5, and is fixed in 10.1.2 and 11.0.0-next.6.

The disclosure coincides with another high-severity path traversal flaw in jsPDF (CVE-2025-68428, CVSS 9.2), patched in version 4.0.0, which could expose arbitrary local files in Node.js environments.

5. RondoDox Botnet Exploits Critical React2Shell Flaw to Hijack IoT Devices and Web Servers

Researchers have uncovered a nine-month-long campaign targeting IoT devices and web applications to build the RondoDox botnet. As of December 2025, attackers are exploiting React2Shell (CVE-2025-55182, CVSS 10.0), a critical flaw in React Server Components and Next.js that enables unauthenticated remote code execution, according to CloudSEK.

Shadowserver estimates 90,000+ instances remain vulnerable worldwide, with the majority in the U.S. RondoDox, active since early 2025, has expanded by abusing multiple N-day flaws, including CVE-2023-1389 and CVE-2025-24893. The campaign evolved from manual scanning to large-scale automated exploitation.

Recent attacks scan for vulnerable Next.js servers and deploy crypto miners, botnet loaders, and a Mirai variant. One tool aggressively removes competing malware and establishes persistence via cron jobs. Mitigations include patching Next.js, isolating IoT devices, deploying WAFs, monitoring suspicious processes, and blocking known C2 infrastructure.

12/17/2025-12/24/2025 Critical n8n Flaw (CVSS 9.9) Enables Arbitrary Code Execution, New UEFI Flaw Enables Early-Boot DMA Attacks And More

1. Critical n8n Flaw (CVSS 9.9) Enables Arbitrary Code Execution Across Thousands of Instances

A critical security vulnerability has been disclosed in the n8n workflow automation platform that could allow arbitrary code execution under certain conditions. The flaw, tracked as CVE-2025-68613, has a CVSS score of 9.9  It has approximately 57,000 weekly downloads on npm. According to the maintainers, expressions provided by authenticated users during workflow configuration may be evaluated in an execution context that is not properly isolated from the underlying runtime. An authenticated attacker could exploit this behavior to execute arbitrary code with the privileges of the n8n process, potentially leading to full system compromise, including data theft, workflow manipulation, and system-level operations. The vulnerability affects all versions from 0.211.0 up to but not including 1.120.4 and has been patched in 1.120.4, 1.121.1, and 1.122.0. Users are strongly urged to update immediately or restrict workflow permissions and harden deployments if patching is delayed.

2. U.S. CISA Adds a Flaw in WatchGuard Fireware OS to its Known Exploited Vulnerabilities Catalog

CISA has added a critical WatchGuard Firebox OS vulnerability, CVE-2025-14733 (CVSS 9.3), to its KEV catalog after active exploitation was confirmed. The flaw is an out-of-bounds write issue in WatchGuard Fireware OS that can be exploited remotely and without authentication via exposed IKEv2 VPN services. When Mobile User VPN or Branch Office VPN with IKEv2 is configured using a dynamic gateway peer, specially crafted network traffic can trigger improper memory handling, allowing attackers to execute arbitrary code on affected Firebox devices. The vulnerability impacts multiple Fireware OS branches, including 11.10.2–11.12.4_Update1, 12.0–12.11.5, and 2025.1–2025.1.3, putting VPN gateways at risk of full compromise.   WatchGuard has released patches, Indicators of Attack, and mitigation guidance. CISA has ordered federal agencies to remediate the flaw by December 26, 2025. Organizations are strongly urged to apply updates immediately, rotate secrets after patching, and restrict exposure if fixes cannot be deployed at once.

3. New UEFI Flaw Enables Early-Boot DMA Attacks on ASRock, ASUS, GIGABYTE, MSI Motherboards

Certain motherboard models from ASRock, ASUS, GIGABYTE, and MSI are affected by a firmware vulnerability that exposes systems to early-boot DMA attacks despite UEFI and IOMMU protections being enabled. Discovered by Nick Peterson and Mohamed Al-Sharifi of Riot Games, the flaw stems from firmware incorrectly reporting that DMA protection is active while failing to initialize the IOMMU during early boot.

This gap allows a malicious PCIe DMA-capable device with physical access to read or modify system memory before the operating system and its security controls load, potentially enabling pre-boot code injection and undermining system integrity. CERT/CC warns attackers could access sensitive data or alter the system’s initial state.

The issue affects multiple Intel and AMD chipset families and is tracked under CVE-2025-14304, CVE-2025-11901, CVE-2025-14302, and CVE-2025-14303, each with a CVSS score of 7.0. Vendors are releasing firmware updates to fix IOMMU initialization. Users are strongly advised to apply patches promptly, especially in environments where physical access cannot be fully controlled.

4. Exploited SonicWall Zero-Day Patched (CVE-2025-40602)

SonicWall has released a hotfix for a local privilege escalation vulnerability, CVE-2025-40602, affecting Secure Mobile Access (SMA) 1000 appliances and warned that the flaw has been exploited in the wild. The vulnerability was reportedly chained with CVE-2025-23006 to achieve unauthenticated remote code execution with root privileges. CVE-2025-23006, patched in January 2025, is a deserialization of untrusted data flaw in the Appliance and Central Management Consoles that allows unauthenticated attackers to execute OS commands. The newly disclosed CVE-2025-40602 also impacts the Appliance Management Console and, due to missing authorization checks, enables attackers with local access to escalate privileges to root when chained with the earlier bug.

SonicWall credited researchers from Google’s Threat Intelligence Group for reporting the issue, though no indicators of compromise have been shared. Customers are urged to upgrade to 12.4.3-03245 or 12.5.0-02283 and restrict management interface access. Even if earlier patches are applied, deploying the latest updates remains essential to fully mitigate the risk.

12/10/2025-12/17/2025 New React RSC Vulnerabilities, Hackers Exploit GitHub, New PCPcat Exploiting React2Shell Vulnerability And More

1. New React RSC Vulnerabilities Enable DoS and Source Code Exposure

The React team has released fixes for newly discovered flaws in React Server Components (RSC) that could lead to denial-of-service (DoS) attacks or source code exposure. The issues were uncovered by security researchers while probing patches for CVE-2025-55182, a critical RSC vulnerability that has already been exploited in the wild. Two vulnerabilities, CVE-2025-55184 and CVE-2025-67779 (both CVSS 7.5), enable pre-authentication DoS through unsafe deserialization that can trigger infinite loops and hang server processes. A third issue, CVE-2025-55183 (CVSS 5.3), may allow attackers to retrieve Server Function source code via crafted HTTP requests under specific conditions. The flaws affect multiple 19.x versions of react-server-dom packages. Researchers RyotaK, Shinsaku Nomura, and Andrew MacPherson reported the issues. Users are strongly advised to upgrade to versions 19.0.3, 19.1.4, or 19.2.3 to mitigate risk.

2. Hackers Exploit GitHub with Fake Repos to Spread PyStoreRAT Malware

Hackers are abusing GitHub by creating fake repositories that impersonate OSINT, GPT, and DeFi tools to distribute PyStoreRAT, a modular remote access trojan designed for data theft and system control. Active since mid-2025, the campaign primarily targets cybersecurity professionals, developers, and cryptocurrency users who trust open-source platforms for tooling. The attackers publish seemingly legitimate Python or JavaScript projects, often promoted on X and YouTube, and artificially inflate stars and forks to build credibility. After users run the code, hidden loaders fetch HTA files from remote servers, ultimately installing PyStoreRAT. In many cases, malicious code is injected later through “maintenance” commits, allowing repositories to appear benign for weeks or months.

Once deployed, PyStoreRAT enables credential and wallet theft, keylogging, and remote command execution while using obfuscation and encrypted communications to evade detection. Researchers warn this campaign highlights growing supply-chain risks on GitHub and recommend strict repository verification, behavioral monitoring, and isolated testing environments as key defenses.

3. FreePBX Patches Critical SQLi, File-Upload, and AUTHTYPE Bypass Flaws Enabling RCE

 Multiple critical vulnerabilities have been disclosed in the FreePBX platform, including a severe authentication bypass flaw.

Discovered by Horizon3.ai, the three primary flaws are:

• CVE-2025-61675 (8.6): Authenticated SQL injections across four endpoints.
• CVE-2025-61678 (8.6): An authenticated file upload flaw allowing PHP web shell deployment.
• CVE-2025-66039 (9.3): An authentication bypass when “Authorization Type” is set to “webserver,” enabling attackers to log into the admin panel with a forged header and insert malicious users.

These easily exploitable issues permit remote code execution. Updates have been released: CVE-2025-61675/61678 are fixed in versions 16.0.92/17.0.6, and CVE-2025-66039 in 16.0.44/17.0.23.

As mitigation, FreePBX advises setting “Authorization Type” to “usermanager” and disabling “Override Readonly Settings.” The “webserver” auth type is now considered legacy and offers reduced security; its configuration option has been removed from the UI. Users should analyze systems where it was enabled for signs of compromise.

4. New PCPcat Exploiting React2Shell Vulnerability to compromise 59,000+ Servers

A new malware campaign dubbed PCPcat has compromised more than 59,000 servers in under 48 hours by exploiting critical vulnerabilities in Next.js and React environments. The attacks abuse two flaws—CVE-2025-29927 and CVE-2025-66478—that enable unauthenticated remote code execution through prototype pollution and command injection.

PCPcat scans public-facing Next.js applications at scale, testing around 2,000 targets per batch every 30–60 minutes, and has achieved an unusually high success rate of 64.6%. Once a vulnerable server is identified, the malware extracts environment files, cloud credentials, SSH keys, and command histories, exfiltrating the data via simple HTTP requests. The operation is coordinated through a command-and-control server in Singapore using three ports: 666 for payload delivery, 888 for reverse tunnels, and 5656 for core management. To maintain persistence, PCPcat installs proxy and tunneling tools, allowing attackers to retain access even after patches are applied.

5. Fortinet Firewalls Under Active Attack

Threat actors are actively exploiting two critical authentication bypass flaws in Fortinet FortiGate devices, tracked as CVE-2025-59718 and CVE-2025-59719 (CVSS 9.8), which affect FortiOS and other products. These vulnerabilities allow attackers to bypass single sign-on protections using crafted SAML messages when FortiCloud SSO is enabled. This feature is enabled by default during FortiCare registration, leaving many organizations unknowingly exposed. In observed attacks, malicious SSO logins from specific hosting providers have been used to gain administrative access, export full device configurations, and steal hashed credentials. Although these hashes require cracking, weak or reused passwords remain vulnerable. CISA has added CVE-2025-59718 to its Known Exploited Vulnerabilities catalog, mandating federal agencies to patch by December 23rd, 2025. To mitigate risk, organizations should immediately apply updates, change all passwords, and restrict management interface access to trusted internal networks only.