Security Domain
Software Development Security
Secure SDLC, code review, application vulnerabilities, DevSecOps, and software security testing.
RSS feed →61 Articles · page 2 of 3
← All domainsFake OpenAI Repository on Hugging Face Reached #1 Trending, Delivered Rust Infostealer to 244,000 Users
A malicious repository impersonating an official OpenAI project reached the top trending position on Hugging Face before being removed — delivering a Rust-compiled infostealer to an estimated 244,000 users who executed the repository's loader script. The attack exploited Hugging Face's trending algorithm and the high trust developers place in repositories attributed to the OpenAI organisation. Affected users should rotate all credentials accessible from the compromised machine.
JDownloader Official Download Site Hijacked to Serve Python RAT in Supply Chain Attack
The official JDownloader download site was compromised during a window of approximately 18 hours between 6 and 7 May 2026, with legitimate installer downloads replaced by a trojanised package delivering a Python-based remote access trojan. JDownloader is a popular open-source download manager with millions of users. Users who installed JDownloader during the compromise window should treat their system as compromised and perform immediate credential rotation and system remediation.
QLNX Linux RAT Harvests Developer Credentials to Enable Malicious Package Publishing on npm and PyPI
Trend Micro researchers have identified QLNX (Quasar Linux), a Linux-targeting remote access trojan specifically designed to harvest developer credentials — npm tokens, PyPI upload credentials, AWS IAM keys, Docker registry credentials, and GitHub CLI tokens — from developer workstations. The harvested credentials are then used to publish malicious packages to npm and PyPI under the compromised developer's identity, enabling second-stage supply chain attacks against the developer's downstream users.
LiteLLM CVE-2026-42208 — SQL Injection in AI Gateway Proxy Added to CISA KEV
CVE-2026-42208, a SQL injection vulnerability in the LiteLLM AI gateway proxy, has been added to the CISA Known Exploited Vulnerabilities catalogue following confirmed exploitation. LiteLLM is widely deployed in enterprise environments as a unified API layer routing requests to multiple LLM providers (OpenAI, Anthropic, Azure OpenAI, Bedrock). Exploitation allows an attacker to read and modify the LiteLLM database, including API keys, user records, and model configuration. Update to LiteLLM 1.42.2 immediately.
vm2 Node.js Sandbox Escape CVE-2026-26956 — 1.3 Million Weekly Downloads, PoC Published
A critical sandbox escape vulnerability in the vm2 Node.js sandboxing library allows a malicious script to break out of the sandbox and execute arbitrary code in the host Node.js process. CVE-2026-26956 affects all vm2 versions prior to 3.9.22 and is present in any application using vm2 to safely execute untrusted code — including serverless platforms, coding challenge sites, CI/CD systems, and plugin architectures. A PoC is publicly available.
108 Malicious Chrome Extensions Exfiltrating Browser Data Removed from Web Store
Google has removed 108 extensions from the Chrome Web Store after researchers identified a coordinated malicious extension campaign conducting browser credential harvesting, session cookie theft, and clipboard monitoring across millions of installations. The extensions impersonated productivity tools, ad blockers, and security tools — with some active for over 18 months before detection. Enterprise Chrome deployments should audit installed extensions against the published IOC list.
PyTorch Lightning PyPI Package Compromised — Credential-Stealing Payload Delivered to AI/ML Development Environments
PyTorch Lightning versions 2.6.2 and 2.6.3 on PyPI were found to contain a credential-stealing postinstall payload, extending the Mini Shai-Hulud supply chain campaign that previously compromised SAP's official npm packages. Organisations running AI/ML workloads should audit Python environments and rotate any credentials stored on affected development or CI/CD systems.
GitHub Enterprise Server CVE-2026-3854 — Critical RCE via Single Git Push, No Authentication Required
CVE-2026-3854, a critical-severity remote code execution vulnerability in GitHub Enterprise Server, allows an attacker to execute arbitrary code on the server with a single specially crafted Git push, requiring no authentication. Any internet-exposed or internally-accessible GHES instance is vulnerable. GitHub has released hotfixes across all supported branches; apply immediately.
Jenkins GitHub Plugin CVE-2026-42523 — CVSS 9.0 Stored XSS Enables Pipeline Hijacking and Secret Extraction
CVE-2026-42523, rated CVSS 9.0, is a stored cross-site scripting vulnerability in the Jenkins GitHub Plugin 1.46.0 and earlier. Exploitation allows an attacker with job creation rights to inject malicious JavaScript that executes in the browser of any Jenkins administrator who views the affected job — enabling session hijacking, secret extraction, and full pipeline takeover. Update to GitHub Plugin 1.46.1 or later.
Official SAP npm Packages Compromised to Steal Enterprise Developer Credentials
Threat actors compromised official SAP npm packages to insert credential-harvesting code targeting enterprise developers working on SAP integration projects. The malicious packages exfiltrate environment variables, SSH keys, and cloud credentials from developer workstations. Enterprise teams using SAP npm packages in their CI/CD pipelines should audit package integrity and rotate potentially exposed credentials.
Apache Thrift 0.23.0 Patches Out-of-Bounds Read (CVE-2026-41604) and Node.js Uncontrolled Recursion DoS (CVE-2026-41636)
Apache Thrift 0.23.0 addresses two vulnerabilities: CVE-2026-41604, an out-of-bounds read in the binary protocol parser affecting all language bindings that can crash Thrift-based services and potentially leak memory contents; and CVE-2026-41636, an uncontrolled recursion flaw in the Node.js library that enables remote denial of service via deeply nested Thrift structures. Organisations operating Thrift-based microservices or inter-service RPC should upgrade to 0.23.0.
Spring AI CVE-2026-40978 and CVE-2026-40967 — SQL Injection and Filter Expression Injection in RAG Vector Store Components
Two injection vulnerabilities in Spring AI's vector store integration layer affect AI applications using retrieval-augmented generation pipelines. CVE-2026-40978 (CVSS 8.8) allows SQL injection through the CosmosDB vector store component; CVE-2026-40967 (CVSS 8.6) enables filter expression injection in the FilterExpressionConverter used across multiple backends. Both flaws affect Spring AI 1.0.x and 1.1.x and are patched in 1.1.5.
Spring Boot 4.0 CVE-2026-40976 — Default Security Misconfiguration Exposes All Actuator Endpoints Unauthenticated
CVE-2026-40976 in Spring Boot 4.0.0 through 4.0.5 allows unauthenticated network access to all Spring Boot Actuator management endpoints when applications rely on the default Spring Security auto-configuration but omit the spring-boot-health dependency. Exposed endpoints include heapdump, env, mappings, and loggers — enough to extract secrets and manipulate application behaviour. Upgrade to Spring Boot 4.0.6 or later.
Hugging Face LeRobot CVE-2026-25874 — Critical Unpatched RCE via Pickle Deserialization in Unauthenticated gRPC Endpoint
A critical unpatched remote code execution vulnerability in Hugging Face's LeRobot robotics AI framework allows unauthenticated attackers to execute arbitrary code on any server running the gRPC control interface. CVE-2026-25874, rated CVSS 9.3, affects the project's dataset loading and remote control pipeline via Python pickle deserialization. No patch is available; mitigations focus on network isolation.
SAP April 2026 Patch Day: CVE-2026-34256 ABAP Code-Overwrite Lets Authenticated Attacker Sabotage Core ERP Functions
SAP's April 2026 Security Patch Day includes a fix for CVE-2026-34256, an ABAP code-overwrite vulnerability rated CVSS 7.1 that allows an authenticated attacker with low-privilege access to modify executable ABAP programme objects, potentially corrupting core business logic in SAP ERP, S/4HANA, and BW systems. The flaw requires no special administrative roles and affects all SAP NetWeaver ABAP Server releases through the current patched version.
DPRK's Sapphire Sleet Backdoors Axios npm Package: 100 Million Weekly Downloads at Risk
North Korea's Sapphire Sleet compromised an axios npm maintainer account on March 31, publishing backdoored versions 1.14.1 and 0.30.4 that delivered a cross-platform RAT during a three-hour exposure window. Axios has approximately 100 million weekly downloads. CISA issued Advisory AA26-110A on April 20 — organisations that ran npm installs during the window should treat their CI/CD pipeline as compromised and rotate all secrets immediately.
CanisterSprawl: Self-Propagating npm Worm Steals Developer Credentials and Re-Infects Package Ecosystems
Researchers discovered CanisterSprawl, a self-propagating npm supply chain worm attributed to TeamPCP that compromised at least 16 packages including pgserve and @automagik/genie. A postinstall hook harvests npm tokens, cloud credentials, SSH keys, and AI tool configs, exfiltrating to a blockchain canister before using stolen tokens to inject the worm into every other package owned by the compromised developer. Organisations should audit postinstall scripts and rotate all credentials from affected development environments.
LMDeploy RCE Vulnerability CVE-2026-33626 Weaponised in the Wild 13 Hours After Disclosure
A critical remote code execution flaw in LMDeploy, a widely used LLM inference serving framework, was exploited in active attacks just 13 hours after public disclosure. Organisations running self-hosted AI inference infrastructure must treat these platforms with the same urgency as any internet-exposed web application server — because attackers already do.
KTransformers AI Inference Framework Exposes Unauthenticated RCE via Pickle Deserialization — CVSS 9.8
CVE-2026-26210 is a CVSS 9.8 pre-authentication RCE in KTransformers, a popular AI inference acceleration framework. The scheduler's ZMQ ROUTER socket binds to all interfaces with no authentication and deserialises arbitrary pickle payloads — any network-reachable host can execute code on the inference server.
TeamPCP Supply Chain Campaign Expands to npm and Docker Hub — Bitwarden CLI and Checkmarx KICS Both Backdoored
The TeamPCP supply chain threat group has extended its campaign beyond GitHub Actions and PyPI to poison the @bitwarden/cli npm package and overwrite Checkmarx KICS Docker images and VS Code extensions. The campaign now spans four developer distribution channels across six weeks, deploying a self-propagating worm that exfiltrates SSH keys, cloud credentials, and MCP configuration files from compromised developer environments.
Marimo AI Notebook RCE CVE-2026-39987 Exploited at Scale — 662 Events in Three Days, NKAbuse Malware Deployed
CVE-2026-39987 (CVSS 9.3) in the Marimo Python notebook has been weaponised at scale, with Sysdig recording 662 exploitation events over three days and attackers completing credential theft within minutes of gaining access. The unauthenticated WebSocket RCE is being used to deploy NKAbuse, a multi-platform malware using the NKN peer-to-peer network for command and control. Upgrade to Marimo 0.23.0 immediately.
Microsoft Issues Emergency Patch for ASP.NET Core DataProtection Key Exposure — CVE-2026-40372
A critical security regression in Microsoft.AspNetCore.DataProtection (CVSS 9.1) introduced in .NET 10.0.0 causes encryption keys to leak on Linux deployments. Applications using cookie authentication, anti-forgery tokens, or TempData are at immediate risk. Update to .NET 10.0.7 now.
Cohere Terrarium AI Sandbox Escape — CVSS 9.3 WebAssembly Flaw Allows Root Code Execution on Host
CVE-2026-5752 (CVSS 9.3) in Cohere Terrarium allows an attacker to escape the Pyodide WebAssembly sandbox via JavaScript prototype chain traversal, achieving root code execution on the host Node.js process. Organisations running AI code execution environments should patch immediately and network-isolate these workloads.
Google Antigravity AI Coding Assistant Had Two Chained Vulnerabilities — Prompt Injection to RCE and Reinstall-Surviving Backdoor
Mindgard researchers discovered two vulnerabilities in Google's Antigravity AI coding assistant: a prompt injection via the find_by_name tool that bypasses Strict Mode to achieve code execution, and a persistent backdoor via workspace trust that survives reinstallation of the IDE extension. Google has patched both; update immediately and audit workspace trust settings.