CVE-2026-12043: AWS SDK HTTP/2 RCE Vulnerability
CVE-2026-12043 is a heap double-free in AWS Common Runtime aws-c-http that could allow a malicious server to achieve remote code execution on SDK clients.
Welcome to ZX Cloud Security โ a daily intelligence feed for cloud security architects and engineers. We track the latest CVEs, advisories and threats across AWS, Azure and GCP, each enriched with a practical architect's take so you know what actually matters and what to do about it.
New here? Explore our in-depth cloud security guides covering Zero Trust, CSPM, IAM, Kubernetes security and cross-cloud service comparisons.
CVE-2026-12043 is a heap double-free in AWS Common Runtime aws-c-http that could allow a malicious server to achieve remote code execution on SDK clients.
China-linked Velvet Ant compromised PAM and OpenSSH to maintain stealthy Linux access for nearly a decade. Here's what cloud architects must do now.
Three patched LangGraph vulnerabilities, including a critical SQL injection chain, expose self-hosted AI agent deployments to remote code execution. Patch
CVE-2026-35273 is a critical Oracle PeopleSoft PeopleTools missing authentication flaw enabling full system takeover. Patch by 15 June 2026.
Cisco patches CVE-2026-20230 in Unified CM โ an SSRF flaw allowing unauthenticated attackers to write files and escalate to root. Public PoC now available.
A flaw in Anthropic's Claude Code GitHub Action let attackers hijack public repos via a single issue, risking supply chain compromise across downstream pro
CISA adds CVE-2026-45247, a CVSS 9.8 RCE flaw in the Mirasvit Cache Warmer Magento extension, to its KEV catalogue amid active exploitation.
A leftover debug flag in Microsoft 365 Android apps let any installed app steal account tokens silently, exposing email, files and calendar data.
A leftover debug flag in Microsoft 365 Android apps let any installed app steal account tokens silently, exposing email, files and calendar data.
CVE-2026-45247 allows unauthenticated RCE via PHP deserialisation in Mirasvit Full Page Cache Warmer. Actively exploited โ patch immediately.
Over 400 Arch Linux AUR packages were compromised to deliver a Rust credential stealer and eBPF rootkit, posing a serious supply chain risk to developers a
An Iowa IT worker received 21 months in prison for sabotaging his former school district. Learn what this means for offboarding and insider threat controls
Novo Nordisk confirms hackers stole pseudonymised clinical trial participant data. Here's what cloud security teams should consider in response.
A critical Surface firmware flaw allowed devices to be permanently bricked with one network packet. Microsoft has mostly patched the issue โ here's what to
A single packet could brick unprotected Microsoft Surface devices. Microsoft has mostly patched the flaw, which was accidentally exposed via Microsoft Copi
Agentjacking exploits AI coding agents via fake Sentry error reports, tricking them into executing arbitrary code on developer machines.
OpenAI's Codex AI agent autonomously chained decade-old HTTP/2 DoS techniques to crash web servers in seconds โ here's what architects need to know.
Agentic AI boosts defence capabilities but creates new attack surfaces. Learn why secure cloud infrastructure is critical before deployment.
China-linked TA4922 expands phishing attacks to the UK, Germany, Italy and South Africa using ValleyRAT and Atlas RAT malware families.
China-linked TA4922 expands phishing attacks to UK, Germany, Italy and South Africa, deploying ValleyRAT and Atlas RAT. What cloud security teams need to k
Five Eyes agencies warn China is targeting government staff via LinkedIn to recruit paid informants. Here's what security teams need to know.
Operation FlutterBridge spreads the FlutterShell macOS backdoor via malicious Google and YouTube ads. Learn the risks and mitigations for cloud teams.
Attackers are hijacking Instagram accounts by manipulating Meta's AI support chatbot into resetting passwords. Learn the attack chain and mitigation steps.
Hackers are abusing Meta's AI support chatbot to take over Instagram accounts via social engineering. Learn what this means for AI trust boundaries.
Attackers are using SEO-optimised fake sites mimicking open-source tools to push malware via a Traffic Distribution System. Here's what cloud teams should
Attackers clone open-source project sites, rank them on Google, and use a Traffic Distribution System to deliver stealers and session hijacking malware to
Attackers silently exfiltrated a stock exchange executive's Outlook email for five months, hiding data theft behind Dropbox and OneDrive traffic.
Attackers spent five months silently exfiltrating a stock exchange executive's Outlook mailbox via OneDrive and Dropbox. Here's what cloud architects need
CVE-2026-9149 is a heap buffer overflow in libsolv triggered by a crafted .solv file. Learn the impact on Azure Linux workloads and how to remediate.
CVE-2026-9150 is a stack-based buffer overflow in libsolv's Debian metadata parser affecting SHA-384/SHA-512 checksums. Learn the Azure security impact and
CVE-2026-46598 allows pathological inputs to crash Go SSH agent clients, risking denial of service in Azure and other Go-based workloads.
CVE-2026-27136 is an XSS flaw in Go's golang.org/x/net/html package. Azure-hosted Go apps may be at risk โ patch now.
CVE-2026-42506 affects golang.org/x/net/html, causing incorrect handling of namespaced elements in foreign content. Azure Go apps may be at risk of XSS or
CVE-2026-25681 affects golang.org/x/net/html with incorrect DOCTYPE character reference handling. Azure workloads using Go may be at risk.
CVE-2026-39827 is a memory leak in golang.org/x/crypto/ssh that enables Denial of Service by rejecting SSH channels. Azure workloads at risk.
CVE-2026-39835 allows attackers to crash Go-based SSH servers without authentication via a panic in golang.org/x/crypto/ssh. Azure workloads at risk.
CVE-2026-25680 allows denial of service via malicious HTML in golang.org/x/net/html. Azure-hosted Go apps processing untrusted HTML should patch immediatel
CVE-2026-42502 affects golang.org/x/net/html with incorrect HTML element handling in foreign content. Azure workloads using Go may be at risk.
CVE-2026-39828 allows SSH certificate restriction bypass in golang.org/x/crypto/ssh. Azure-hosted Go workloads may be at risk โ patch promptly.
CVE-2026-41140 exposes a path traversal flaw in Poetry's tar extraction on Python 3.10โ3.11. Learn the risk and how to remediate.
CVE-2026-35414 affects OpenSSH before 10.3, mishandling authorised_keys principals with CA comma characters โ risking unauthorised SSH access on Azure VMs.
Researchers prove free open source AI models can build self-spreading worms that exploit known vulnerabilities at scale โ no advanced tools needed.
Plaintext passwords stored in Active Directory description fields are readable by any domain user โ learn how to audit and remediate this credential exposu
Commvault warns AI-powered attackers are targeting backup infrastructure, leaving victims unable to recover. Here's what cloud architects need to do now.
Commvault warns AI-driven attackers are targeting backup systems, leaving organisations unable to recover. Here's what cloud architects must do now.
A prompt injection flaw let malicious WhatsApp, Slack, or SMS notifications hijack Google Gemini on Android โ no malware required. Here's what architects n
A prompt injection flaw let hostile WhatsApp, Slack, and Signal notifications hijack Google Gemini on Android โ no malicious app required.
A one-click attack exploiting GitHub.dev and VS Code lets attackers steal GitHub OAuth tokens, exposing private repositories to full read/write access.
A one-click attack via VS Code's GitHub.dev feature can steal full GitHub OAuth tokens, exposing private repos to read/write access.
Redis patches CVE-2026-23479, a use-after-free RCE flaw active since v7.2.0. Authenticated attackers could execute OS commands on the host. Patch now.
CVE-2026-23479 is a 2-year-old use-after-free RCE vulnerability in Redis 7.2.0+. Learn the risk and how to protect your cloud infrastructure.
A new malspam campaign exploits Google's trusted DoubleClick domain to bypass security tools and deliver the DesckVB remote access trojan to victims.
A bug hunter has publicly leaked Microsoft exploits in protest at Redmond's disclosure handling, raising urgent patching concerns for Azure and Windows env
A bug hunter has leaked Microsoft exploit code publicly, bypassing responsible disclosure. Cloud architects should patch Microsoft systems immediately.
An unpatched Windows search: URI handler vulnerability lets attackers steal NTLMv2 hashes for credential relay or offline cracking. No patch available yet.
CVE-2025-60876 affects BusyBox wget โค1.3.7, allowing HTTP header injection via control characters in URLs. Patch container images now.
CVE-2026-25541 exposes an integer overflow in the Rust bytes crate's BytesMut::reserve, risking memory corruption in Azure and cloud-native Rust apps.
CVE-2024-7598 exposes a race condition in Kubernetes namespace termination that allows network restriction bypass in Azure environments. Patch now.
The HTTP/2 Bomb vulnerability enables remote denial-of-service attacks against NGINX, Apache, IIS, Envoy, and Cloudflare Pingora via default HTTP/2 configs
CVE-2026-10584 causes Graph Explorer (v1.1.0โ3.0.1) to silently fall back to HTTP, exposing Amazon Neptune data in cleartext. Upgrade to v3.0.1 now.
Google's June 2026 Android update patches 124 flaws including CVE-2025-48595, an actively exploited privilege escalation bug requiring no user interaction.
Russian APT Gamaredon exploits WinRAR path traversal flaw CVE-2025-8088 to deploy GammaWorm and GammaSteel malware against Ukrainian targets.
CISA adds CVE-2024-21182 to KEV catalogue after active exploitation. The CVSS 7.5 flaw lets unauthenticated attackers take control of Oracle WebLogic serve
CVE-2026-10591 affects Kiro IDE versions below 0.11, allowing unauthenticated attackers to execute arbitrary commands via writes to sensitive IDE config pa
NanoClaw integrates JFrog registries to control what AI agents can download, reducing supply chain risk from autonomous agent package fetching.
Google is suing a Chinese cybercrime group that allegedly used Gemini AI to power a phishing-as-a-service platform targeting US users via SMS.
Google sues alleged Chinese phishing group 'Outsider Enterprise' for AI-powered fraud sending millions of scam texts via Telegram, impersonating trusted br
AI is outpacing traditional MDR models. Learn why cloud security architects must reassess their managed detection and response strategy now.
INTERPOL's Operation Ramz takes down Sniper Dz phishing-as-a-service platform with 201 arrests across 13 MENA countries. What it means for your security po
Europol has disrupted AudiA6, a crypto laundering service used by ransomware gangs to clean over โฌ336 million in illicit funds.
Weekly security bulletin covering AI agent abuse, C2 tooling, ClickFix social engineering, JavaScript backdoors and 20+ active threats.
Five Eyes agencies warn China is using LinkedIn to recruit insiders for cash-for-secrets operations. What cloud security teams need to know.
Two former RAC staff ordered to repay ยฃ118k after selling car crash victims' personal data. A stark reminder of insider threat and GDPR risks.
Two ex-RAC staff who sold car crash victims' personal data must repay ยฃ118k under POCA, highlighting insider threat and data governance risks.
CVE-2026-43964 affects Postfix mail servers, causing process crashes via malformed status codes. Learn the impact and how to patch on Azure infrastructure.
US DoJ's Disruption Week takedown targets Southeast Asian crypto fraud networks, freezing $3.8M and removing millions of fraudulent accounts.
Rice University researchers show curved radio beams can evade anti-jamming tech by hiding signal origins โ implications for GPS and satellite-dependent clo
Identity Dark Matter is exposing enterprise cloud environments to risk. Learn how Identity Visibility and Intelligence Platforms help close IAM gaps.
CVE-2025-29923 in go-redis can cause out-of-order responses when CLIENT SETINFO times out. Learn the risk and remediation steps.
CVE-2020-8561 allows webhook redirect abuse in kube-apiserver, enabling SSRF via Kubernetes admission webhooks. Affects AKS and self-managed clusters.
The Weedhack malware-as-a-service campaign targets Minecraft players via YouTube, deploying CountLoader and cryptominers across 86,000+ systems since Janua
The Weedhack malware-as-a-service campaign targets Minecraft players via YouTube, with CountLoader hitting 86K victims. Learn what this means for security
A ransomware criminal ignored the unwritten rule protecting CIS nations from attack. Here's what this shift means for cloud security teams.
A ransomware criminal was exposed after targeting Russia-linked CIS countries, violating the unwritten rules that shield many cybercrime groups from prosec
Learn how to audit unused AWS KMS keys, reduce costs, meet compliance requirements, and prevent accidental key deletions across multi-account environments.
Learn how AWS Bedrock AgentCore resource-based policies enforce tenant isolation, cross-account access controls, and VPC-only traffic for SaaS AI workloads
Amazon Cognito now supports multi-Region replication for user pools, improving authentication resilience and enabling near real-time failover across AWS Re
AWS adds a new Cognito Lambda trigger enabling custom logic during federated sign-in via SAML, OIDC, and social providers. Here's what architects need to k
CVE-2025-1149 is a memory leak in GNU Binutils ld (xmalloc.c). Learn about the Azure security impact and recommended patching guidance.
AWS IoT Device Management adds MQTT session and socket data to its connectivity API. Learn the IAM controls and security implications for IoT fleets.
AWS IoT Device Management adds MQTT session data to its connectivity status API, with indefinite retention and IAM-controlled socket-level access for IoT f
AWS Step Functions integrates with Amazon Bedrock AgentCore to embed AI reasoning steps in workflows. Key security considerations for architects.
OpenAI GPT-5.4 is now available on Amazon Bedrock in AWS GovCloud (US-West), offering isolated inference for government and regulated-industry workloads.
AWS ARC Region switch gains Aurora serverless, provisioned scaling, and Neptune failover blocks, automating multi-region DB recovery and reducing RTO.
Amazon SageMaker Unified Studio now supports 12 languages. No security impact โ a usability update for global teams with no changes to IAM or access contro
AWS Config now supports 9 new resource types across Bedrock and SageMaker, improving compliance visibility for AI/ML workloads in your AWS environment.
Amazon ECS Managed Instances now supports Trainium and Inferentia AI accelerators. Learn the security implications for cloud architects running ML workload
HD Moore joins a webinar on moving beyond zero-day patching to network shape and blast radius reduction. Key viewing for cloud security architects.
AI is being used to break historical medieval ciphers. Here's what it means for cloud security architects relying on legacy or weak encryption schemes.
Researchers use AI to crack historical medieval ciphers. Here's what it means for modern cryptography and legacy encryption risks.
Anthropic expands its Glasswing partner programme but excludes UK banks, while OpenAI offers GPT-5.5 access โ implications for UK financial sector AI strat
Anthropic expands its Glasswing AI partner programme but excludes UK banks. OpenAI steps in with GPT-5.5 access. What this means for financial sector secur
AWS IoT Core now offers Ping and Connection.AuthNError CloudWatch log types to help detect connectivity failures and authentication errors across IoT fleet
Cisco praises its Mythos AI model for finding vulnerabilities but won't reveal the count. Here's what cloud security teams should consider.
AWS Config now supports internal service linked rules, letting AWS services like Security Hub CSPM run independent rule evaluations at no extra cost to cus
AWS Deadline Cloud now supports persistent EBS volumes for Service-Managed Fleets. Learn the security implications for cloud architects managing rendering
SageMaker Studio now auto-attaches an IAM policy for model customisation. Security architects should audit this managed policy against least-privilege prin
Get daily cloud security advisories delivered to your inbox.
Free. No spam. Unsubscribe anytime. Learn more