Hello, stay ahead with CISO Brief 🚀

🔒 IBM patched critical vulnerabilities in Langflow, an open-source framework for building AI agents and LLM-powered workflows. The flaws could allow attackers to bypass authentication, leak API credentials, and execute arbitrary code, and have been exploited previously. Two issues are rated critical (9.8 CVSS); fixes are in Langflow 1.13.0 available on PyPI and Docker. IBM also published upgrade guidance and Hono patches.

📧 Amazon Connect Customer now lets agents pause and resume email interactions while awaiting information or approvals, enabling them to take higher-priority work in the meantime. Paused duration is tracked separately from active handling time, giving supervisors clearer metrics. Agents can manage pause/resume from the Agent Workspace or Contact Control Panel, and supervisors can monitor paused emails and configure flows to auto-resume or transfer after thresholds.

🛡️ The NCSC warns AI-enabled malware is “highly likely” to be used in ransomware attacks against UK critical infrastructure within two years. Dr. Richard Horne cautioned that underinvestment and technical debt leave organisations exposed, while AI will amplify attackers’ speed and scale rather than invent wholly new techniques. The report highlights state-linked groups, ongoing router and OT targeting, and four primary attack vectors driving most incidents.

🔔 Customers of the mSecure password manager received an email on November 3 from developer mSeven Software warning of "suspicious activity" linked to the app and urging users to consider changing passwords, resetting 2FA and removing passkeys. The notice offered few specifics: it did not disclose what systems were affected, whether vaults or encryption keys were accessed, how many users were notified, or whether the activity is ongoing. mSeven said it has launched an investigation, limited the issue, and engaged external cybersecurity experts. The company asserts vault data is encrypted on-device prior to sync, though a 2025 BSI study found that an older Android build (mSecure 6.1.5) could, in theory, allow developer access to stored data.

🔒 Nammu21 and Google Cloud developed a proof-of-concept that converts bespoke syndicated loan agreements into executable, on‑ledger contracts using the Google Cloud Universal Ledger (GCUL). The platform renders legal terms as Digital Credit Instruments (DCIs) and instantiates them as gculpy contracts to automate onboarding, transfers, payments, and consent-driven workflows. This approach aims to reduce settlement times, improve transparency, and maintain strict permissioning and privacy for regulated institutions.

🔍 Microsoft Threat Intelligence describes a fake-download campaign that uses trojanized Inno Setup-style installers to deliver multiple monetization payloads during a single installation session. The installers track per-offer callbacks and execution outcomes, enabling delivery of proxyware, distributed-compute, and adware-style utilities without informed user consent. The post details execution chains, indicators, MITRE mappings, Defender coverage, and mitigation guidance for defenders.

🤖 This article from Google Cloud Consulting outlines five practical lessons for enterprises adopting agentic AI systems. It emphasizes redesigning workflows from relay-like handoffs to parallelized, agent-driven processes and shifting humans from step-level validators to outcome owners. The piece highlights trust-building through delegation practices, clear responsibilities, and low-risk experiments, while recommending hub-and-spoke governance and cross-team collaboration to accelerate safe, scalable adoption.

🛠️ Today Google AI & Infrastructure and Mercor introduce the Physical Engineers' Last Exam (PLE), a new benchmark and open competition to test agents on mechanical, electrical, civil, and structural engineering tasks. PLE provides 29 public tasks, grading tools, and a leaderboard at ple-bench.com, and includes long-horizon, real-world project folders with diverse file types. The benchmark evaluates code compliance, safety, and cost efficiency, revealing a wide gap between frontier models and professional engineering requirements.

🛡️ Microsoft was named a Leader in the inaugural 2026 Gartner Magic Quadrant for AI Agent Management Platforms, reflecting momentum behind Agent 365. The platform applies identity-first controls with Microsoft Entra Agent ID, integrates with existing Microsoft security and management tools, and supports a broad ecosystem of agents from vendors and open-source frameworks. Microsoft is expanding discovery, policy coverage, and automated governance features such as the new Agent 365 Action Center.

📧 Amazon Connect Customer now auto-fills the customer’s email address from the linked customer profile when agents create a new email thread from a case. This reduces manual entry and the risk of errors, saving agents time. The Cases feature is available in multiple AWS regions, including US, Canada, Europe, Asia Pacific, and Africa.

🔔 AWS Config now supports 10 additional AWS resource types across services including Amazon ElastiCache and Amazon QuickSight. The update enables better discovery, assessment, auditing, and remediation across a broader set of resources. If recording for all resource types is enabled, AWS Config will automatically begin tracking these new types. The new resource types are also available for use in Config rules and Config aggregators.

🔒 Barracuda disclosed two vulnerabilities—a pre-auth authorization bypass (CVE-2026-116606) and a command injection (CVE-2026-111560)—that can be chained to bypass controls and gain root on affected devices. CVE-2026-116606 affects Email Security Gateway, while CVE-2026-111560 impacts multiple products including Load Balancer ADC, Message Archiver, WAF Control Center, Web Application Firewall, and Web Security Gateway. Exploitation requires access to the administrative interface; organizations are advised to restrict admin access to trusted networks, apply network access controls, and install the latest firmware and patches. Multiple-version-outdated firmware may need sequential updates.

🛡️ Google warns that threat actors are increasingly hijacking enterprise access to Large Language Models (LLMs) to fraudulently consume costly compute and AI services. Attackers obtain API keys via phishing, leaked code, misconfigurations, or exposed tokens to run expensive workloads while victims incur the charges. Incidents span multiple regions and industries, with single-day fraudulent charges exceeding $100,000 and cumulative losses reaching millions in some cases.

🛡️ BTCPay Server reported that attackers stole cryptocurrency from hot wallets on instances running version 2.4.1 or earlier, with about $1.63m traced to linked Bitcoin and Litecoin addresses. Blockchain analysis by Bitquery identified repeated sweeps between October 30 and November 2, and some activity continued after operators were warned to shut down. The project urged affected operators to rebuild servers with new wallets and seeds and noted that hardware wallets and wallets with spending keys held off-server were not impacted.

📰 Zero Wire is a new public vulnerability intelligence feed on Cloudflare Radar designed to give security teams concise, actionable briefings on critical, internet-facing vulnerabilities. Each card summarizes the issue, affected deployments, recommended actions, and links to vendor advisories and evidence. Briefings include contextual network observations when available, a suggested agent prompt for impact assessment, and an update chronology to track developments. The service complements vendor advisories and vulnerability databases rather than cataloging every CVE.

🛡️ Google outlines expanded measures to counter rising scam losses across the Asia Pacific region, which reached up to $114.1 billion USD in 2025. The company is deploying built-in protections such as Fake Call Detection and Pixel scam alerts, expanding Android developer verification, and promoting RCS-based defenses. Google is also engaging in industry coordination through the Industry Accord and new alliances like JASA, while Google.org commits $6 million to local partners for community training, AI prototyping, and policy support.

🔐 Late October’s emergency PHP patches for CVE-2026-109618 and CVE-2026-109619 aimed to stop remote code execution that could affect roughly 70% of websites. Despite fixes for multiple PHP branches, Censys found millions of vulnerable servers still unpatched two weeks later, while exploitation attempts and new exploits, including a related Brotli issue, continued to appear. Organizations must update PHP, audit systems, and watch for intrusion signs.

🧊 Supermarkets in at least nine European countries reported a cyber-attack that manipulated refrigeration systems, spoiling chilled and frozen stock. Retailers in Belgium, Estonia, Finland, Ireland, Italy, Luxembourg, the Netherlands, Portugal and Sweden were affected, with temperatures reportedly set as high as 50°C. Authorities including Sweden’s NCSC and EU officials are investigating, while some UK food-supply businesses have also been impacted. The incidents mirror an October attack on refrigeration providers and raise concerns about deliberate disruption of food supply chains and potential nation-state involvement.