{"month":"2025-08","headlines":[{"id":"34d2cf20a680","published":"2025-08-31T18:35:00.000Z","title":"Brokewell Android malware delivered through fake TradingView ads","url":"https://bleepingcomputer.com/news/security/brokewell-android-malware-delivered-through-fake-tradingview-ads","summary":"Cybercriminals are abusing Meta's advertising platforms with fake offers of a free TradingView Premium app that spreads the Brokewell malware for Android.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"f58fee36cb5b","published":"2025-08-31T14:49:48.000Z","title":"VULSOLVER: Vulnerability Detection via LLM-Driven Constraint Solving","url":"https://arxiv.org/abs/2509.00882","summary":"Traditional vulnerability detection methods rely heavily on predefined rule matching, which often fails to capture vulnerabilities accurately. With the rise of large language models (LLMs), leveraging their ability to understand code semantics has emerged as a promising…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"e9b933c870ea","published":"2025-08-31T08:42:00.000Z","title":"The Cost of AI Velocity: 5 Actions Dev Leaders Must Take to Secure Their Codebase From AI Vulnerabilities","url":"https://checkmarx.com/blog/the-cost-of-ai-velocity-5-actions-dev-leaders-must-take-to-secure-their-codebase-from-ai-vulnerabilities","summary":"AI-powered coding tools accelerate development but introduce alarming vulnerability rates. This blog reveals the 5 essential actions dev leaders must take to safeguard codebases against AI-driven security risks.","feed":"checkmarx","publisher":"checkmarx","topics":[],"type":"analysis"},{"id":"acf189e1bb75","published":"2025-08-31T08:00:21.000Z","title":"Week in review: 300k+ Plex Media Server instances still vulnerable to attack, exploited Git RCE flaw","url":"https://helpnetsecurity.com/2025/08/31/week-in-review-300k-plex-media-server-instances-still-vulnerable-to-attack-exploited-git-rce-flaw","summary":"Here’s an overview of some of last week’s most interesting news, articles, interviews and videos: 300k+ Plex Media Server instances still vulnerable to attack via CVE-2025-34158 Over 300,000 internet-facing Plex Media Server instances are still vulnerable to attack via…","feed":"helpnetsecurity","publisher":"helpnetsecurity","topics":[],"type":"roundup"},{"id":"b8df39f07805","published":"2025-08-31T01:20:58.000Z","title":"Wrap Up: The Month of AI Bugs","url":"https://embracethered.com/blog/posts/2025/wrapping-up-month-of-ai-bugs","summary":"That’s it. The Month of AI Bugs is done. There won’t be a post tomorrow, because I will be at PAX West. Overview of Posts ChatGPT: Exfiltrating Your Chat History and Memories With Prompt Injection | Video ChatGPT Codex: Turning ChatGPT Codex Into a ZombAI Agent | Video…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","ai-security"],"type":"analysis"},{"id":"803c8f139f41","published":"2025-08-30T16:22:00.000Z","title":"TamperedChef infostealer delivered through fraudulent PDF Editor","url":"https://bleepingcomputer.com/news/security/tamperedchef-infostealer-delivered-through-fraudulent-pdf-editor","summary":"Threat actors have been using multiple websites promoted through Google ads to distribute a convincing PDF editing app that delivers an info-stealing malware called TamperedChef.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"d16f2a0879fe","published":"2025-08-30T12:11:23.000Z","title":"Partially Functional Dynamic Backdoor Diffusion-based Causal Model","url":"https://arxiv.org/abs/2509.00472","summary":"Causal inference in spatio-temporal settings is critically hindered by unmeasured confounders with complex spatio-temporal dynamics and the prevalence of multi-resolution data. While diffusion models present a promising avenue for estimating structural causal models, existing…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"009f61b2b218","published":"2025-08-30T12:06:00.003Z","title":"Attackers Abuse Velociraptor Forensic Tool to Deploy Visual Studio Code for C2 Tunneling","url":"https://thehackernews.com/2025/08/attackers-abuse-velociraptor-forensic.html","summary":"Cybersecurity researchers have called attention to a cyber attack in which unknown threat actors deployed an open-source endpoint monitoring and digital forensic tool called Velociraptor, illustrating ongoing abuse of legitimate software for malicious purposes. \"In this…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"research"},{"id":"7b7c65bcdc53","published":"2025-08-30T07:04:29.000Z","title":"The Resurgence of GCG Adversarial Attacks on Large Language Models","url":"https://arxiv.org/abs/2509.00391","summary":"Gradient-based adversarial prompting, such as the Greedy Coordinate Gradient (GCG) algorithm, has emerged as a powerful method for jailbreaking large language models (LLMs). In this paper, we present a systematic appraisal of GCG and its annealing-augmented variant, T-GCG,…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"43bd21879770","published":"2025-08-30T06:35:32.000Z","title":"Backdoor Samples Detection Based on Perturbation Discrepancy Consistency in Pre-trained Language Models","url":"https://arxiv.org/abs/2509.05318","summary":"The use of unvetted third-party and internet data renders pre-trained models susceptible to backdoor attacks. Detecting backdoor samples is critical to prevent backdoor activation during inference or injection during training. However, existing detection methods often require…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"4ad5292d7f8d","published":"2025-08-30T04:36:00.005Z","title":"WhatsApp Patches Zero-Click Exploit Targeting iOS and macOS Devices","url":"https://thehackernews.com/2025/08/whatsapp-issues-emergency-update-for.html","summary":"WhatsApp has addressed a security vulnerability in its messaging apps for Apple iOS and macOS that it said may have been exploited in the wild in conjunction with a recently disclosed Apple flaw in targeted zero-day attacks. The vulnerability, CVE-2025-55177 (CVSS score: 5.4),…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"analysis"},{"id":"4e01ba4889f2","published":"2025-08-30T03:20:58.000Z","title":"AgentHopper: An AI Virus","url":"https://embracethered.com/blog/posts/2025/agenthopper-a-poc-ai-virus","summary":"As part of the Month of AI Bugs, serious vulnerabilities that allow remote code execution via indirect prompt injection were discovered. There was a period of a few weeks where multiple arbitrary code execution vulnerabilities existed in popular agents, like GitHub Copilot,…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","ai-security"],"type":"analysis"},{"id":"157bf0e96b1f","published":"2025-08-29T22:06:22.000Z","title":"GHSA-g5qg-72qw-gw5v / CVE-2025-57752: Next.js Affected by Cache Key Confusion for Image Optimization API Routes","url":"https://github.com/advisories/GHSA-g5qg-72qw-gw5v","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"1c24a80fdd3b","published":"2025-08-29T21:59:55.000Z","title":"GHSA-xv57-4mr9-wg8v / CVE-2025-55173: Next.js Content Injection Vulnerability for Image Optimization","url":"https://github.com/advisories/GHSA-xv57-4mr9-wg8v","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6d0c7b936eb9","published":"2025-08-29T21:33:09.000Z","title":"GHSA-4342-x723-ch2f / CVE-2025-57822: Next.js Improper Middleware Redirect Handling Leads to SSRF","url":"https://github.com/advisories/GHSA-4342-x723-ch2f","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ef33d357eb09","published":"2025-08-29T21:32:02.000Z","title":"GHSA-876g-49r6-33qj / CVE-2025-43773: Liferay Portal allows improper access through the expandoTableLocalService","url":"https://github.com/advisories/GHSA-876g-49r6-33qj","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"25ecd3b242dd","published":"2025-08-29T20:33:56.000Z","title":"GHSA-xwfj-jgwm-7wp5 / CVE-2025-58160:  Tracing logging user input may result in poisoning logs with ANSI escape sequences","url":"https://github.com/advisories/GHSA-xwfj-jgwm-7wp5","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"49203cb8eb78","published":"2025-08-29T20:31:17.000Z","title":"GHSA-6h9x-9j5v-7w9h / CVE-2024-52284: Rancher Fleet Helm Values are stored inside BundleDeployment in plain text","url":"https://github.com/advisories/GHSA-6h9x-9j5v-7w9h","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"3f7f6a79cb8a","published":"2025-08-29T20:30:04.000Z","title":"GHSA-9q78-27f3-2jmh: webp crate may expose memory contents when encoding an image","url":"https://github.com/advisories/GHSA-9q78-27f3-2jmh","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"cb0acc8a9e5c","published":"2025-08-29T20:23:16.000Z","title":"GHSA-82ff-hg59-8x73 / CVE-2025-47909: github.com/gorilla/csrf improperly validates TrustedOrigins allowing CSRF attacks","url":"https://github.com/advisories/GHSA-82ff-hg59-8x73","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6ddf098b4a8d","published":"2025-08-29T20:20:13.000Z","title":"GHSA-9fvj-xqr2-xwg8 / CVE-2025-58157: gnark affected by denial of service when computing scalar multiplication using fake-GLV algorithm","url":"https://github.com/advisories/GHSA-9fvj-xqr2-xwg8","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"1cc68134ba17","published":"2025-08-29T20:08:24.000Z","title":"GHSA-hw6f-rjfj-j7j7 / CVE-2025-58068: Eventlet affected by HTTP request smuggling in unparsed trailers","url":"https://github.com/advisories/GHSA-hw6f-rjfj-j7j7","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca","dast"],"type":"advisory"},{"id":"f641926eae88","published":"2025-08-29T20:07:47.000Z","title":"GHSA-5jch-xhw4-r43v / CVE-2025-58067: Google Sign-In for Rails allowed redirect to protocol-relative URI","url":"https://github.com/advisories/GHSA-5jch-xhw4-r43v","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"472e56d77814","published":"2025-08-29T20:07:27.000Z","title":"GHSA-4855-q42w-5vr4 / CVE-2025-58066: DoS Vulnerability in ntpd-rs","url":"https://github.com/advisories/GHSA-4855-q42w-5vr4","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d89f1759146d","published":"2025-08-29T17:22:00.002Z","title":"Researchers Warn of Sitecore Exploit Chain Linking Cache Poisoning and Remote Code Execution","url":"https://thehackernews.com/2025/08/researchers-warn-of-sitecore-exploit.html","summary":"Three new security vulnerabilities have been disclosed in the Sitecore Experience Platform that could be exploited to achieve information disclosure and remote code execution. The flaws, per watchTowr Labs, are listed below - CVE-2025-53693 - HTML cache poisoning through unsafe…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"research"},{"id":"f4c98620539b","published":"2025-08-29T16:41:57.000Z","title":"GHSA-w469-hj2f-jpr5 / CVE-2025-58158: Harness Allows Arbitrary File Write in Gitness LFS server","url":"https://github.com/advisories/GHSA-w469-hj2f-jpr5","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d563e4ef3fd9","published":"2025-08-29T16:31:00.000Z","title":"WhatsApp patches vulnerability exploited in zero-day attacks","url":"https://bleepingcomputer.com/news/security/whatsapp-patches-vulnerability-exploited-in-zero-day-attacks","summary":"WhatsApp has patched a security vulnerability in its iOS and macOS messaging clients that was exploited in targeted zero-day attacks.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"advisory"},{"id":"ad86e565770f","published":"2025-08-29T16:24:12.000Z","title":"GHSA-v2ch-c8v8-fgr7: Versity panic induced by AWS chunked data sent to port","url":"https://github.com/advisories/GHSA-v2ch-c8v8-fgr7","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"559272a54208","published":"2025-08-29T16:14:42.000Z","title":"Under the hood: Exploring the AI models powering GitHub Copilot","url":"https://github.blog/ai-and-ml/github-copilot/under-the-hood-exploring-the-ai-models-powering-github-copilot","summary":"Learn how GitHub Copilot’s evolving models and infrastructure center developer choice and power agentic workflows.","feed":"github-ai","publisher":"github","topics":["agentic-sdlc"],"type":"analysis"},{"id":"2a6c26f02d54","published":"2025-08-29T15:38:23.000Z","title":"GHSA-4h45-jpvh-6p5j / CVE-2024-58259: Rancher affected by unauthenticated Denial of Service","url":"https://github.com/advisories/GHSA-4h45-jpvh-6p5j","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"fa378b1fb00f","published":"2025-08-29T15:34:37.000Z","title":"GHSA-hq8m-v68g-8cf8 / CVE-2025-55202: Opencast has a partial path traversal vulnerability in UI config","url":"https://github.com/advisories/GHSA-hq8m-v68g-8cf8","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"1dde3da744c2","published":"2025-08-29T15:30:38.000Z","title":"GHSA-694p-3fxc-m92h / CVE-2025-9654: AiondaDotCom mcp-ssh command injection vulnerability in SSH operations","url":"https://github.com/advisories/GHSA-694p-3fxc-m92h","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca","ai-security"],"type":"advisory"},{"id":"4e106c4344f4","published":"2025-08-29T14:59:37.000Z","title":"GHSA-m54q-mm9w-fp6g / CVE-2025-55304: Exiv2 has quadratic performance in ICC profile parsing in JpegBase::readMetadata","url":"https://github.com/advisories/GHSA-m54q-mm9w-fp6g","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d5384346f482","published":"2025-08-29T14:49:41.000Z","title":"GHSA-496f-x7cq-cq39 / CVE-2025-54080: Exiv2 Segmentation Faults in Exiv2::EpsImage::writeMetadata() via crafted EPS file","url":"https://github.com/advisories/GHSA-496f-x7cq-cq39","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"81126dbf50a0","published":"2025-08-29T14:32:48.000Z","title":"Cybersecurity AI: Hacking the AI Hackers via Prompt Injection","url":"https://arxiv.org/abs/2508.21669","summary":"We demonstrate how AI-powered cybersecurity tools can be turned against themselves through prompt injection attacks. Prompt injection is reminiscent of cross-site scripting (XSS): malicious text is hidden within seemingly trusted content, and when the system processes it, that…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"1914596de611","published":"2025-08-29T14:21:00.000Z","title":"Microsoft says recent Windows update didn't kill your SSD","url":"https://bleepingcomputer.com/news/microsoft/microsoft-says-recent-KB5063878-windows-update-didnt-kill-your-ssd","summary":"Microsoft has found no link between the August 2025 KB5063878 security update and customer reports of failure and data corruption issues affecting solid-state drives (SSDs) and hard disk drives (HDDs).","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"analysis"},{"id":"3e1f7c374fe6","published":"2025-08-29T14:00:06.000Z","title":"Detecting Stealthy Data Poisoning Attacks in AI Code Generators","url":"https://arxiv.org/abs/2508.21636","summary":"Deep learning (DL) models for natural language-to-code generation have become integral to modern software development pipelines. However, their heavy reliance on large amounts of data, often collected from unsanitized online sources, exposes them to data poisoning attacks,…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast","models"],"type":"research"},{"id":"572e2a8fca7c","published":"2025-08-29T13:51:28.000Z","title":"Human-Written vs. AI-Generated Code: A Large-Scale Study of Defects, Vulnerabilities, and Complexity","url":"https://arxiv.org/abs/2508.21634","summary":"As AI code assistants become increasingly integrated into software development workflows, understanding how their code compares to human-written programs is critical for ensuring reliability, maintainability, and security. In this paper, we present a large-scale comparison of…","feed":"arxiv-agentic","publisher":"arxiv","topics":["agentic-sdlc"],"type":"research"},{"id":"16df71224377","published":"2025-08-29T13:00:49.000Z","title":"Show HN: A minimal TS library that generates prompt injection attacks","url":"https://prompt-injector.blueprintlab.io/","summary":"Hacker News: 33 points, 15 comments.","feed":"hn-ai-security-0","publisher":"hacker-news","topics":["ai-security"],"type":"analysis"},{"id":"6a688286b832","published":"2025-08-29T12:32:35.000Z","title":"Agentic Discovery and Validation of Android App Vulnerabilities","url":"https://arxiv.org/abs/2508.21579","summary":"Existing Android vulnerability detection tools overwhelm teams with thousands of low-signal warnings yet uncover few true positives. Analysts spend days triaging these results, creating a bottleneck in the security pipeline. Meanwhile, genuinely exploitable vulnerabilities…","feed":"arxiv-crse","publisher":"arxiv","topics":[],"type":"research"},{"id":"3c1a02f866ba","published":"2025-08-29T12:31:11.000Z","title":"GHSA-26rv-h2hf-3fw4 / CVE-2025-4644: Payload's SQLite adapter Session Fixation vulnerability","url":"https://github.com/advisories/GHSA-26rv-h2hf-3fw4","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"4c074e015f73","published":"2025-08-29T12:31:11.000Z","title":"GHSA-5v66-m237-hwf7 / CVE-2025-4643: Payload does not invalidate JWTs after log out","url":"https://github.com/advisories/GHSA-5v66-m237-hwf7","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"fcee8328a26c","published":"2025-08-29T10:30:00.000Z","title":"Can Your Security Stack See ChatGPT? Why Network Visibility Matters","url":"https://thehackernews.com/2025/08/can-your-security-stack-see-chatgpt-why.html","summary":"Generative AI platforms like ChatGPT, Gemini, Copilot, and Claude are increasingly common in organizations. While these solutions improve efficiency across tasks, they also present new data leak prevention for generative AI challenges. Sensitive information may be shared…","feed":"thehackernews","publisher":"thehackernews","topics":["agentic-sdlc"],"type":"analysis"},{"id":"e4b0670ed993","published":"2025-08-29T09:58:00.001Z","title":"Click Studios Patches Passwordstate Authentication Bypass Vulnerability in Emergency Access Page","url":"https://thehackernews.com/2025/08/click-studios-patches-passwordstate.html","summary":"Click Studios, the developer of enterprise-focused password management solution Passwordstate, said it has released security updates to address an authentication bypass vulnerability in its software. The high-severity issue, which is yet to be assigned a CVE identifier, has…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"analysis"},{"id":"87f91cbb570a","published":"2025-08-29T09:44:00.005Z","title":"FreePBX Servers Targeted by Zero-Day Flaw, Emergency Patch Now Available","url":"https://thehackernews.com/2025/08/freepbx-servers-targeted-by-zero-day.html","summary":"The Sangoma FreePBX Security Team has issued an advisory warning about an actively exploited FreePBX zero-day vulnerability that impacts systems with an administrator control panel (ACP) exposed to the public internet. FreePBX is an open-source private branch exchange (PBX)…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"announcement"},{"id":"c3a0eecbc2bf","published":"2025-08-29T08:38:58.000Z","title":"An Empirical Study of Vulnerable Package Dependencies in LLM Repositories","url":"https://arxiv.org/abs/2508.21417","summary":"Large language models (LLMs) have developed rapidly in recent years, revolutionizing various fields. Despite their widespread success, LLMs heavily rely on external code dependencies from package management systems, creating a complex and interconnected LLM dependency supply…","feed":"arxiv-sca","publisher":"arxiv","topics":["sca"],"type":"research"},{"id":"88d34d7ef015","published":"2025-08-29T08:14:52.000Z","title":"A Whole New World: Creating a Parallel-Poisoned Web Only AI-Agents Can See","url":"https://arxiv.org/abs/2509.00124","summary":"This paper introduces a novel attack vector that leverages website cloaking techniques to compromise autonomous web-browsing agents powered by Large Language Models (LLMs). As these agents become more prevalent, their unique and often homogenous digital fingerprints -…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"51a03017421e","published":"2025-08-29T06:00:19.000Z","title":"New framework aims to outsmart malware evasion tricks","url":"https://helpnetsecurity.com/2025/08/29/erdalt-malware-detection-framework","summary":"Attackers have learned how to trick machine learning malware detectors with small but clever code changes, and researchers say they may finally have an answer. In a new paper, academics from Inria and the CISPA Helmholtz Center for Information Security describe a framework that…","feed":"helpnetsecurity","publisher":"helpnetsecurity","topics":["models"],"type":"standard"},{"id":"3e9230159270","published":"2025-08-29T05:00:40.000Z","title":"AI isn’t taking over the world, but here’s what you should worry about","url":"https://helpnetsecurity.com/2025/08/29/ai-threats-explained-video","summary":"In this Help Net Security video, Josh Meier, Senior Generative AI Author at Pluralsight, debunks the myth that AI could “escape” servers or act on its own. He explains how large language models actually work, why they can’t become sentient, and where the real risks lie: from…","feed":"helpnetsecurity","publisher":"helpnetsecurity","topics":[],"type":"analysis"},{"id":"0713d3a9490a","published":"2025-08-29T04:30:50.000Z","title":"AI can’t deliver without trusted, well-governed information","url":"https://helpnetsecurity.com/2025/08/29/ai-readiness-complexity","summary":"While enterprise IT leaders recognize the transformative potential of AI, a gap in information readiness is causing their organizations to struggle in securing, governing, and aligning AI initiatives across business, according to a survey conducted by the Ponemon Institute. Who…","feed":"helpnetsecurity","publisher":"helpnetsecurity","topics":["trust"],"type":"analysis"},{"id":"fcecdb1f9ddc","published":"2025-08-29T04:00:57.000Z","title":"Infosec products of the month: August 2025","url":"https://helpnetsecurity.com/2025/08/29/infosec-products-of-the-month-august-2025","summary":"Here’s a look at the most interesting products from the past month, featuring releases from: Black Kite, Brivo, Cloudflare, Descope, Doppel, Druva, Elastic, ExtraHop, LastPass, Prove, Riverbed, Rubrik, StackHawk, and Trellix. StackHawk empowers security teams to expand their…","feed":"helpnetsecurity","publisher":"helpnetsecurity","topics":["dast"],"type":"analysis"},{"id":"94b81d67ab7c","published":"2025-08-29T01:47:07.000Z","title":"Locus: Agentic Predicate Synthesis for Directed Fuzzing","url":"https://arxiv.org/abs/2508.21302","summary":"Directed fuzzing aims to find program inputs that lead to specified target program states. It has broad applications, such as debugging system crashes, confirming reported bugs, and generating exploits for potential vulnerabilities. This task is inherently challenging because…","feed":"arxiv-dast","publisher":"arxiv","topics":["dast"],"type":"research"},{"id":"8116762d7a79","published":"2025-08-29T00:00:00.000Z","title":"CVE-2025-57819: Sangoma FreePBX Authentication Bypass Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2025-57819","summary":"Sangoma FreePBX contains an authentication bypass vulnerability due to insufficiently sanitized user-supplied data allows unauthenticated access to FreePBX Administrator leading to arbitrary database manipulation and remote code execution. US federal agencies must act by 2025-09-","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"206f410aba85","published":"2025-08-28T21:31:26.000Z","title":"GHSA-8f82-53h8-2p34 / CVE-2025-6203: HashiCorp Vault Community Edition Denial of Service Though Complex JSON Payloads","url":"https://github.com/advisories/GHSA-8f82-53h8-2p34","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"403d5ad819cf","published":"2025-08-28T19:36:22.000Z","title":"GHSA-jc7w-c686-c4v9 / CVE-2025-58058: github.com/ulikunitz/xz leaks memory when decoding a corrupted multiple LZMA archives","url":"https://github.com/advisories/GHSA-jc7w-c686-c4v9","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"2816286f9e1e","published":"2025-08-28T19:36:04.000Z","title":"GHSA-3rw9-wmc8-8948: Coder accepts an APIKey beyond the linked OIDC expiry if there is no refresh token","url":"https://github.com/advisories/GHSA-3rw9-wmc8-8948","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"7411d5b286d9","published":"2025-08-28T19:20:58.000Z","title":"Windsurf MCP Integration: Missing Security Controls Put Users at Risk","url":"https://embracethered.com/blog/posts/2025/windsurf-dangers-lack-of-security-controls-for-mcp-server-tool-invocation","summary":"Part of my default test cases for coding agents is to check how MCP integration looks like, especially if the agent can be configured to allow setting fine-grained controls for tools. Sometimes there are basic security controls missing. Especially when running an agent on your…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","ai-security"],"type":"analysis"},{"id":"5ee764908920","published":"2025-08-28T18:46:43.000Z","title":"High-severity vulnerability in Passwordstate credential manager. Patch now.","url":"https://arstechnica.com/security/2025/08/high-severity-vulnerability-in-passwordstate-credential-manager-patch-now","summary":"Vulnerability can be exploited to gain access to customers' crown jewels.","feed":"ars-security","publisher":"ars-security","topics":["remediation"],"type":"analysis"},{"id":"3b4f3b7a11f6","published":"2025-08-28T17:10:00.002Z","title":"Researchers Find VS Code Flaw Allowing Attackers to Republish Deleted Extensions Under Same Names","url":"https://thehackernews.com/2025/08/researchers-find-vs-code-flaw-allowing.html","summary":"Cybersecurity researchers have discovered a loophole in the Visual Studio Code Marketplace that allows threat actors to reuse names of previously removed extensions. Software supply chain security outfit ReversingLabs said it made the discovery after it identified a malicious…","feed":"thehackernews","publisher":"thehackernews","topics":["sca"],"type":"research"},{"id":"c029478f85ab","published":"2025-08-28T17:08:51.000Z","title":"How we accelerated Secret Protection engineering with Copilot","url":"https://github.blog/ai-and-ml/github-copilot/how-we-accelerated-secret-protection-engineering-with-copilot","summary":"Learn how the Secret Protection engineering team collaborated with GitHub Copilot coding agent to expand validity check coverage.","feed":"github-ai","publisher":"github","topics":["agentic-sdlc","remediation"],"type":"analysis"},{"id":"35983f273dc5","published":"2025-08-28T17:08:00.000Z","title":"Malware devs abuse Anthropic’s Claude AI to build ransomware","url":"https://bleepingcomputer.com/news/security/malware-devs-abuse-anthropics-claude-ai-to-build-ransomware","summary":"Anthropic's Claude Code large language model has been abused by threat actors who used it in data extortion campaigns and to develop ransomware packages.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["agentic-sdlc"],"type":"incident"},{"id":"682368371ffd","published":"2025-08-28T16:46:40.000Z","title":"GHSA-vxg3-w9rv-rhr2: Contrast leaks workload secrets to logs on INFO level","url":"https://github.com/advisories/GHSA-vxg3-w9rv-rhr2","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"19f3fadbe4bb","published":"2025-08-28T16:46:10.000Z","title":"GHSA-w48j-pp7j-fj55 / CVE-2025-58059: Valtimo scripting engine can be used to gain access to sensitive data or resources","url":"https://github.com/advisories/GHSA-w48j-pp7j-fj55","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c7b680d3013d","published":"2025-08-28T16:16:00.000Z","title":"Passwordstate dev urges users to patch auth bypass vulnerability","url":"https://bleepingcomputer.com/news/security/passwordstate-dev-urges-users-to-patch-auth-bypass-vulnerability-as-soon-as-possible","summary":"Click Studios, the company behind the Passwordstate enterprise-grade password manager, has warned customers to patch a high-severity authentication bypass vulnerability as soon as possible.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"analysis"},{"id":"6258f5978453","published":"2025-08-28T15:34:28.000Z","title":"GHSA-xjhf-7833-3pm5 / CVE-2025-58047: Volto affected by possible DoS by invoking specific URL by anonymous user","url":"https://github.com/advisories/GHSA-xjhf-7833-3pm5","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"9d4114b59d40","published":"2025-08-28T15:30:42.000Z","title":"GHSA-4fxf-xgrm-8fcj / CVE-2025-56236: FormCms avatar upload feature has a stored cross-site scripting (XSS) vulnerability","url":"https://github.com/advisories/GHSA-4fxf-xgrm-8fcj","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ab8db778ce9d","published":"2025-08-28T15:19:07.000Z","title":"PromptSleuth: Detecting Prompt Injection via Semantic Intent Invariance","url":"https://arxiv.org/abs/2508.20890","summary":"Large Language Models (LLMs) are increasingly integrated into real-world applications, from virtual assistants to autonomous agents. However, their flexibility also introduces new attack vectors-particularly Prompt Injection (PI), where adversaries manipulate model behavior…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"bb74cc09d6ef","published":"2025-08-28T15:10:13.000Z","title":"GHSA-9m7c-m33f-3429 / CVE-2025-58049: XWiki PDF export jobs store sensitive cookies unencrypted in job statuses","url":"https://github.com/advisories/GHSA-9m7c-m33f-3429","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"31c13cc4a31b","published":"2025-08-28T15:00:00.000Z","title":"MATLAB dev says ransomware gang stole data of 10,000 people","url":"https://bleepingcomputer.com/news/security/matlab-dev-says-ransomware-gang-stole-data-of-over-10-000-people","summary":"MathWorks, a leading developer of mathematical simulation and computing software, revealed that a ransomware gang stole the data of over 10,000 people after breaching its network in April.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"b62746a1b094","published":"2025-08-28T14:59:39.000Z","title":"AVIATOR: Towards AI-Agentic Vulnerability Injection Workflow for High-Fidelity, Large-Scale Code Security Dataset","url":"https://arxiv.org/abs/2508.20866","summary":"The increasing complexity of software systems and the sophistication of cyber-attacks have underscored the need for reliable automated software vulnerability detection. Data-driven approaches using deep learning models show promise but critically depend on the availability of…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast","ai-security"],"type":"research"},{"id":"a837ffe35b5b","published":"2025-08-28T14:58:22.000Z","title":"GHSA-qqfq-7cpp-hcqj / CVE-2025-57759: Contao does not properly manage privileges for page and article fields","url":"https://github.com/advisories/GHSA-qqfq-7cpp-hcqj","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b4f912099bf4","published":"2025-08-28T14:58:00.000Z","title":"GHSA-w53m-gxvg-vx7p / CVE-2025-57757: Contao can disclose sensitive information in the news module","url":"https://github.com/advisories/GHSA-w53m-gxvg-vx7p","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"973082545827","published":"2025-08-28T14:57:45.000Z","title":"GHSA-2xmj-8wmq-7475 / CVE-2025-57756: Contao discloses sensitive information in the front end search index","url":"https://github.com/advisories/GHSA-2xmj-8wmq-7475","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"e2123637e5ae","published":"2025-08-28T14:57:04.000Z","title":"Misleading Large Language Models used (or misused) in Scientific Peer-Reviewing via Hidden Prompt-Injection Attacks","url":"https://arxiv.org/abs/2508.20863","summary":"Large Language Models (LLMs) are increasingly being integrated into the scientific peer-review process, raising new questions about their reliability and resilience to manipulation. In this work, we investigate the potential for hidden prompt injection attacks, where authors…","feed":"arxiv-aisec","publisher":"arxiv","topics":["models","ai-security"],"type":"research"},{"id":"7e4b73ad5bd2","published":"2025-08-28T14:40:45.000Z","title":"GHSA-7m47-r75r-cx8v / CVE-2025-57758: Contao applies improper access control in the back end voters","url":"https://github.com/advisories/GHSA-7m47-r75r-cx8v","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"96209a90f350","published":"2025-08-28T14:40:27.000Z","title":"JADES: A Universal Framework for Jailbreak Assessment via Decompositional Scoring","url":"https://arxiv.org/abs/2508.20848","summary":"Accurately determining whether a jailbreak attempt has succeeded is a fundamental yet unresolved challenge. Existing evaluation methods rely on misaligned proxy indicators or naive holistic judgments. They frequently misinterpret model responses, leading to inconsistent and…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"a9e35a73f22a","published":"2025-08-28T14:40:08.000Z","title":"GHSA-65rg-554r-9j5x / CVE-2024-48908: lychee link checking action affected by arbitrary code injection in composite action","url":"https://github.com/advisories/GHSA-65rg-554r-9j5x","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"0424625744ba","published":"2025-08-28T14:14:24.000Z","title":"Multi-Agent Penetration Testing AI for the Web","url":"https://arxiv.org/abs/2508.20816","summary":"AI-powered development platforms are making software creation accessible to a broader audience, but this democratization has triggered a scalability crisis in security auditing. With studies showing that up to 40% of AI-generated code contains vulnerabilities, the pace of…","feed":"arxiv-dast","publisher":"arxiv","topics":["dast","agentic-sdlc"],"type":"research"},{"id":"118e2e07fbde","published":"2025-08-28T13:33:36.000Z","title":"GHSA-8pxw-9c75-6w56 / CVE-2025-8077: NeuVector admin account has insecure default password","url":"https://github.com/advisories/GHSA-8pxw-9c75-6w56","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"e00932d21c28","published":"2025-08-28T13:33:23.000Z","title":"GHSA-w54x-xfxg-4gxq / CVE-2025-54467: NeuVector process with sensitive arguments lead to leakage","url":"https://github.com/advisories/GHSA-w54x-xfxg-4gxq","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"f452e73bc001","published":"2025-08-28T13:33:14.000Z","title":"GHSA-8ff6-pc43-jwv3 / CVE-2025-53884: NeuVector has an insecure password storage and is vulnerable to rainbow attack","url":"https://github.com/advisories/GHSA-8ff6-pc43-jwv3","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c60ea629a569","published":"2025-08-28T13:00:10.000Z","title":"Unpacking Passkeys Pwned: Possibly the most specious research in decades","url":"https://arstechnica.com/security/2025/08/new-research-claiming-passkeys-can-be-stolen-is-pure-nonsense","summary":"Researchers take note: When the endpoint is compromised, all bets are off.","feed":"ars-security","publisher":"ars-security","topics":["sca"],"type":"analysis"},{"id":"cce4e83051d7","published":"2025-08-28T12:23:38.000Z","title":"Agentic AI coding assistant helped attacker breach, extort 17 distinct organizations","url":"https://helpnetsecurity.com/2025/08/28/agentic-ai-malicious-use","summary":"Cybercriminals have started “vibe hacking” with AI’s help, AI startup Anthropic has shared in a report released on Wednesday. An attacker used the agentic AI coding assistant Claude Code for nearly all steps of a data extortion operation that has targeted at least 17…","feed":"helpnetsecurity","publisher":"helpnetsecurity","topics":["agentic-sdlc","ai-security"],"type":"incident"},{"id":"6ff086e31adb","published":"2025-08-28T12:04:57.000Z","title":"TransUnion says hackers stole 4.4 million customers’ personal information","url":"https://techcrunch.com/2025/08/28/transunion-says-hackers-stole-4-4-million-customers-personal-information","summary":"The credit reporting giant confirmed unauthorized access to a third-party application storing the personal information of its customers, including Social Security numbers.","feed":"techcrunch-security","publisher":"techcrunch-security","topics":[],"type":"incident"},{"id":"50de0412067b","published":"2025-08-28T10:45:31.000Z","title":"CyberSleuth: Autonomous Blue-Team LLM Agent for Web Attack Forensics","url":"https://arxiv.org/abs/2508.20643","summary":"Post-mortem analysis of compromised systems is a key aspect of cyber forensics, today a mostly manual, slow, and error-prone task. Agentic AI, i.e., LLM-powered agents, is a promising avenue for automation. However, applying such agents to cybersecurity remains largely…","feed":"arxiv-agentic","publisher":"arxiv","topics":[],"type":"research"},{"id":"bb08b92e2d79","published":"2025-08-28T10:36:00.008Z","title":"Malicious Nx Packages in ‘s1ngularity’ Attack Leaked 2,349 GitHub, Cloud, and AI Credentials","url":"https://thehackernews.com/2025/08/malicious-nx-packages-in-s1ngularity.html","summary":"The maintainers of the nx build system have alerted users to a supply chain attack that allowed attackers to publish malicious versions of the popular npm package and other auxiliary plugins with data-gathering capabilities. \"Malicious versions of the nx package, as well as…","feed":"thehackernews","publisher":"thehackernews","topics":["sca"],"type":"incident"},{"id":"21f10085e54a","published":"2025-08-28T07:13:49.000Z","title":"Cycuity Radix-ST identifies security weaknesses earlier in chip design","url":"https://helpnetsecurity.com/2025/08/28/cycuity-radix-st","summary":"Cycuity introduces Radix-ST, a new product in its Radix portfolio of security solutions designed to leverage static analysis techniques to identify potential weaknesses early in the design cycle without requiring simulation or emulation. As hardware designs become increasingly…","feed":"helpnetsecurity","publisher":"helpnetsecurity","topics":["sast"],"type":"analysis"},{"id":"6f3cb4044b8f","published":"2025-08-28T04:23:44.000Z","title":"MindGuard: Intrinsic Decision Inspection for Securing LLM Agents Against Metadata Poisoning","url":"https://arxiv.org/abs/2508.20412","summary":"The Model Context Protocol (MCP) is increasingly adopted to standardize the interaction between LLM agents and external tools. However, this trend introduces a new threat: Tool Poisoning Attacks (TPA), where tool metadata is poisoned to induce the agent to perform unauthorized…","feed":"arxiv-sca","publisher":"arxiv","topics":["models","ai-security"],"type":"research"},{"id":"c53a83abb479","published":"2025-08-28T00:30:25.000Z","title":"Poison Once, Refuse Forever: Weaponizing Alignment for Injecting Bias in LLMs","url":"https://arxiv.org/abs/2508.20333","summary":"Large Language Models (LLMs) are aligned to meet ethical standards and safety requirements by training them to refuse answering harmful or unsafe prompts. In this paper, we demonstrate how adversaries can exploit LLMs' alignment to implant bias, or enforce targeted censorship…","feed":"arxiv-models","publisher":"arxiv","topics":["trust","models"],"type":"research"},{"id":"925d7227b60b","published":"2025-08-27T22:38:00.000Z","title":"Storm-0501 hackers shift to ransomware attacks in the cloud","url":"https://bleepingcomputer.com/news/security/storm-0501-hackers-shift-to-ransomware-attacks-in-the-cloud","summary":"Microsoft warns that a threat actor tracked as Storm-0501 has evolved its operations, shifting away from encrypting devices with ransomware to focusing on cloud-based encryption, data theft, and extortion.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"9f5c1013ad7f","published":"2025-08-27T21:24:10.000Z","title":"Network-Level Prompt and Trait Leakage in Local Research Agents","url":"https://arxiv.org/abs/2508.20282","summary":"We show that Web and Research Agents (WRAs) -- language-model-based systems that investigate complex topics on the Internet -- are vulnerable to inference attacks by passive network observers. Deployment of WRAs \\emph{locally} by organizations and individuals for privacy,…","feed":"arxiv-agentic","publisher":"arxiv","topics":[],"type":"research"},{"id":"723818d794c0","published":"2025-08-27T20:47:00.000Z","title":"Experimental PromptLock ransomware uses AI to encrypt, steal data","url":"https://bleepingcomputer.com/news/security/experimental-promptlock-ransomware-uses-ai-to-encrypt-steal-data","summary":"Threat researchers discovered the first AI-powered ransomware, called PromptLock, that uses Lua scripts to steal and encrypt data on Windows, macOS, and Linux systems.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"e5cfe8769901","published":"2025-08-27T19:04:00.001Z","title":"Storm-0501 Exploits Entra ID to Exfiltrate and Delete Azure Data in Hybrid Cloud Attacks","url":"https://thehackernews.com/2025/08/storm-0501-exploits-entra-id-to.html","summary":"The financially motivated threat actor known as Storm-0501 has been observed refining its tactics to conduct data exfiltration and extortion attacks targeting cloud environments. \"Unlike traditional on-premises ransomware, where the threat actor typically deploys malware to…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"494984fd5c32","published":"2025-08-27T18:31:55.000Z","title":"GHSA-4x4m-3c2p-qppc / CVE-2025-5187: Kubernetes Nodes can delete themselves by adding an OwnerReference","url":"https://github.com/advisories/GHSA-4x4m-3c2p-qppc","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"4e2468c39c54","published":"2025-08-27T18:31:55.000Z","title":"GHSA-rfh2-8vxq-jqr8 / CVE-2025-50979: NodeBB SQL Injection vulnerability","url":"https://github.com/advisories/GHSA-rfh2-8vxq-jqr8","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"8b1c4be6f88d","published":"2025-08-27T18:31:55.000Z","title":"GHSA-f2m2-4q6r-cwc4 / CVE-2025-51667: simple-admin-core SQL Injection vulnerability","url":"https://github.com/advisories/GHSA-f2m2-4q6r-cwc4","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"1a9d3d7c1d65","published":"2025-08-27T17:48:33.000Z","title":"Quoting Bruce Schneier","url":"https://simonwillison.net/2025/Aug/27/bruce-schneier","summary":"We simply don’t know to defend against these attacks. We have zero agentic AI systems that are secure against these attacks. Any AI that is working in an adversarial environment—and by this I mean that it may encounter untrusted training data or input—is vulnerable to prompt…","feed":"simonw","publisher":"simonw","topics":["ai-security"],"type":"analysis"},{"id":"6f13f411db48","published":"2025-08-27T17:07:00.005Z","title":"Someone Created the First AI-Powered Ransomware Using OpenAI's gpt-oss:20b Model","url":"https://thehackernews.com/2025/08/someone-created-first-ai-powered.html","summary":"Cybersecurity company ESET has disclosed that it discovered an artificial intelligence (AI)-powered ransomware variant codenamed PromptLock. Written in Golang, the newly identified strain uses the gpt-oss:20b model from OpenAI locally via the Ollama API to generate malicious…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"incident"},{"id":"b65b525ab81b","published":"2025-08-27T16:59:54.000Z","title":"Microsoft details Storm-0501’s focus on ransomware in the cloud","url":"https://cyberscoop.com/storm-0501-ransomware-microsoft-threat-intelligence","summary":"The financially motivated threat group demonstrates deep knowledge of hybrid cloud environments, which allows it to rapidly steal sensitive data, destroy backups and encrypt systems for ransomware.","feed":"cyberscoop","publisher":"cyberscoop","topics":[],"type":"incident"},{"id":"351efa8c8391","published":"2025-08-27T16:48:00.000Z","title":"Over 28,000 Citrix devices vulnerable to new exploited RCE flaw","url":"https://bleepingcomputer.com/news/security/over-28-200-citrix-instances-vulnerable-to-actively-exploited-rce-bug","summary":"More than 28,200 Citrix instances are vulnerable to a critical remote code execution vulnerability tracked as CVE-2025-7775 that is already being exploited in the wild.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"f23a03c4c4f8","published":"2025-08-27T16:46:16.000Z","title":"GHSA-7pwc-wh6m-44q3 / CVE-2025-57821: Google Sign-In for Rails allowed redirects to malformed URLs","url":"https://github.com/advisories/GHSA-7pwc-wh6m-44q3","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"403df5f3bf47","published":"2025-08-27T16:42:47.000Z","title":"GHSA-cxm3-wv7p-598c / CVE-2025-10894: Malicious versions of Nx were published","url":"https://github.com/advisories/GHSA-cxm3-wv7p-598c","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-malware","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"eae1c9f827d5","published":"2025-08-27T16:34:53.000Z","title":"Pruning Strategies for Backdoor Defense in LLMs","url":"https://arxiv.org/abs/2508.20032","summary":"Backdoor attacks are a significant threat to the performance and integrity of pre-trained language models. Although such models are routinely fine-tuned for downstream NLP tasks, recent work shows they remain vulnerable to backdoor attacks that survive vanilla fine-tuning.…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"e810fb230ad0","published":"2025-08-27T16:13:22.000Z","title":"Trustify joins GUAC","url":"https://openssf.org/blog/2025/08/27/trustify-joins-guac","summary":"By Ben Cotton and Dejan Bosanac The superpower of open source is multiple people working together on a common goal. That works for projects, too. GUAC and Trustify are two...","feed":"openssf","publisher":"openssf","topics":["sca"],"type":"analysis"},{"id":"79be96067d66","published":"2025-08-27T16:00:00.000Z","title":"Storm-0501’s evolving techniques lead to cloud-based ransomware","url":"https://microsoft.com/en-us/security/blog/2025/08/27/storm-0501s-evolving-techniques-lead-to-cloud-based-ransomware","summary":"Financially motivated threat actor Storm-0501 has continuously evolved their campaigns to achieve sharpened focus on cloud-based tactics, techniques, and procedures (TTPs). While the threat actor has been known for targeting hybrid cloud environments, their primary objective…","feed":"microsoft-security","publisher":"microsoft","topics":[],"type":"incident"},{"id":"4a4fdbec0b93","published":"2025-08-27T15:39:46.000Z","title":"Evaluating Language Model Reasoning about Confidential Information","url":"https://arxiv.org/abs/2508.19980","summary":"As language models are increasingly deployed as autonomous agents in high-stakes settings, ensuring that they reliably follow user-defined rules has become a critical safety concern. To this end, we study whether language models exhibit contextual robustness, or the capability…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"ab510237310d","published":"2025-08-27T15:33:15.000Z","title":"GHSA-9hp3-f5g8-rccg / CVE-2025-52122: The Freeform CraftCMS plugin contains an Server-side template injection (SSTI) vulnerability","url":"https://github.com/advisories/GHSA-9hp3-f5g8-rccg","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"993337a2cb98","published":"2025-08-27T15:20:58.000Z","title":"Cline: Vulnerable To Data Exfiltration And How To Protect Your Data","url":"https://embracethered.com/blog/posts/2025/cline-vulnerable-to-data-exfiltration","summary":"Cline is quite a popular AI coding agent, according to the product website it has 2+ million downloads and over 47k stars on GitHub. Unfortunately, Cline is vulnerable to data exfiltration through the rendering of markdown images from untrusted domains in the chat box. This…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","ai-security"],"type":"guide"},{"id":"de5a4d6600cb","published":"2025-08-27T15:10:00.006Z","title":"Anthropic Disrupts AI-Powered Cyberattacks Automating Theft and Extortion Across Critical Sectors","url":"https://thehackernews.com/2025/08/anthropic-disrupts-ai-powered.html","summary":"Anthropic on Wednesday revealed that it disrupted a sophisticated operation that weaponized its artificial intelligence (AI)-powered chatbot Claude to conduct large-scale theft and extortion of personal data in July 2025. \"The actor targeted at least 17 distinct organizations,…","feed":"thehackernews","publisher":"thehackernews","topics":["ai-security"],"type":"analysis"},{"id":"16b224d6bc91","published":"2025-08-27T15:00:00.000Z","title":"Microsoft ranked number one in modern endpoint security market share third year in a row","url":"https://microsoft.com/en-us/security/blog/2025/08/27/microsoft-ranked-number-one-in-modern-endpoint-security-market-share-third-year-in-a-row","summary":"For a third year a row, Microsoft has been named the number one leader for endpoint security market share, as featured in a new IDC report.","feed":"microsoft-security","publisher":"microsoft","topics":[],"type":"analysis"},{"id":"6957d8ca160e","published":"2025-08-27T12:25:45.000Z","title":"AEGIS : Automated Co-Evolutionary Framework for Guarding Prompt Injections Schema","url":"https://arxiv.org/abs/2509.00088","summary":"Prompt injection attacks pose a significant challenge to the safe deployment of Large Language Models (LLMs) in real-world applications. While prompt-based detection offers a lightweight and interpretable defense strategy, its effectiveness has been hindered by the need for…","feed":"arxiv-aisec","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"74e4409021d5","published":"2025-08-27T12:00:54.000Z","title":"s1ngularity: supply chain attack leaks secrets on GitHub: everything you need to know","url":"https://wiz.io/blog/s1ngularity-supply-chain-attack","summary":"Detect and mitigate a critical supply chain compromise affecting the Nx NPM Package. Organizations should act urgently.","feed":"wiz","publisher":"wiz","topics":["sca"],"type":"incident"},{"id":"f09247d10349","published":"2025-08-27T11:07:59.000Z","title":"We Are Still Unable to Secure LLMs from Malicious Inputs","url":"https://schneier.com/blog/archives/2025/08/we-are-still-unable-to-secure-llms-from-malicious-inputs.html","summary":"Nice indirect prompt injection attack : Bargury’s attack starts with a poisoned document, which is shared to a potential victim’s Google Drive. (Bargury says a victim could have also uploaded a compromised file to their own account.) It looks like an official document on…","feed":"schneier","publisher":"schneier","topics":["ai-security"],"type":"analysis"},{"id":"ad13fe7d3321","published":"2025-08-27T10:00:00.000Z","title":"OpenAI and Anthropic share findings from a joint safety evaluation","url":"https://openai.com/index/openai-anthropic-safety-evaluation","summary":"OpenAI and Anthropic share findings from a first-of-its-kind joint safety evaluation, testing each other’s models for misalignment, instruction following, hallucinations, jailbreaking, and more—highlighting progress, challenges, and the value of cross-lab collaboration.","feed":"openai","publisher":"openai","topics":["models"],"type":"analysis"},{"id":"5b3f46b35b0c","published":"2025-08-27T09:39:00.002Z","title":"Salesloft OAuth Breach via Drift AI Chat Agent Exposes Salesforce Customer Data","url":"https://thehackernews.com/2025/08/salesloft-oauth-breach-via-drift-ai.html","summary":"A widespread data theft campaign has allowed hackers to breach sales automation platform Salesloft to steal OAuth and refresh tokens associated with the Drift artificial intelligence (AI) chat agent. The activity, assessed to be opportunistic in nature, has been attributed to a…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"incident"},{"id":"c0fea56fede7","published":"2025-08-27T09:06:28.000Z","title":"Safety Alignment Should Be Made More Than Just A Few Attention Heads","url":"https://arxiv.org/abs/2508.19697","summary":"Current safety alignment for large language models(LLMs) continues to present vulnerabilities, given that adversarial prompting can effectively bypass their safety measures.Our investigation shows that these safety mechanisms predominantly depend on a limited subset of…","feed":"arxiv-sca","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"eaeb77c311c5","published":"2025-08-27T07:08:44.000Z","title":"Private, Verifiable, and Auditable AI Systems","url":"https://arxiv.org/abs/2509.00085","summary":"The growing societal reliance on artificial intelligence necessitates robust frameworks for ensuring its security, accountability, and trustworthiness. This thesis addresses the complex interplay between privacy, verifiability, and auditability in modern AI, particularly in…","feed":"arxiv-agentic","publisher":"arxiv","topics":["trust"],"type":"research"},{"id":"6d211e3ef50c","published":"2025-08-27T01:11:59.000Z","title":"Servant, Stalker, Predator: How An Honest, Helpful, And Harmless (3H) Agent Unlocks Adversarial Skills","url":"https://arxiv.org/abs/2508.19500","summary":"This paper identifies and analyzes a novel vulnerability class in Model Context Protocol (MCP) based agent systems. The attack chain describes and demonstrates how benign, individually authorized tasks can be orchestrated to produce harmful emergent behaviors. Through…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"ae839c43b86e","published":"2025-08-27T00:18:49.000Z","title":"PoolFlip: A Multi-Agent Reinforcement Learning Security Environment for Cyber Defense","url":"https://arxiv.org/abs/2508.19488","summary":"Cyber defense requires automating defensive decision-making under stealthy, deceptive, and continuously evolving adversarial strategies. The FlipIt game provides a foundational framework for modeling interactions between a defender and an advanced adversary that compromises a…","feed":"arxiv-agentic","publisher":"arxiv","topics":[],"type":"research"},{"id":"75d84c98ce1a","published":"2025-08-27T00:00:00.000Z","title":"The One Where We Just Steal The Vulnerabilities (CrushFTP CVE-2025-54309)","url":"https://labs.watchtowr.com/the-one-where-we-just-steal-the-vulnerabilities-crushftp-cve-2025-54309","summary":"","feed":"watchtowr","publisher":"watchtowr","topics":[],"type":"analysis"},{"id":"e682fcaff74e","published":"2025-08-26T23:23:35.000Z","title":"SIExVulTS: Sensitive Information Exposure Vulnerability Detection System using Transformer Models and Static Analysis","url":"https://arxiv.org/abs/2508.19472","summary":"Sensitive Information Exposure (SIEx) vulnerabilities (CWE-200) remain a persistent and under-addressed threat across software systems, often leading to serious security breaches. Existing detection tools rarely target the diverse subcategories of CWE-200 or provide…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"c9fe09ec856f","published":"2025-08-26T22:43:25.000Z","title":"Piloting Claude for Chrome","url":"https://simonwillison.net/2025/Aug/26/piloting-claude-for-chrome","summary":"Piloting Claude for Chrome Two days ago I said : I strongly expect that the entire concept of an agentic browser extension is fatally flawed and cannot be built safely. Today Anthropic announced their own take on this pattern, implemented as an invite-only preview Chrome…","feed":"simonw","publisher":"simonw","topics":["ai-security"],"type":"analysis"},{"id":"4611bb020a59","published":"2025-08-26T22:33:14.000Z","title":"GHSA-vj54-72f3-p5jv / CVE-2025-57820: devalue prototype pollution vulnerability","url":"https://github.com/advisories/GHSA-vj54-72f3-p5jv","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"0956c929de40","published":"2025-08-26T22:29:31.000Z","title":"Reliable Weak-to-Strong Monitoring of LLM Agents","url":"https://arxiv.org/abs/2508.19461","summary":"We stress test monitoring systems for detecting covert misbehavior in autonomous LLM agents (e.g., secretly sharing private information). To this end, we systematize a monitor red teaming (MRT) workflow that incorporates: (1) varying levels of agent and monitor situational…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"903bbabe84b3","published":"2025-08-26T22:23:00.000Z","title":"Google to verify all Android devs to protect users from malware","url":"https://bleepingcomputer.com/news/security/google-to-verify-all-android-devs-to-protect-users-from-malware","summary":"Google is introducing a new defense for Android called 'Developer Verification' to block malware installations from sideloaded apps sourced from outside the official Google Play app store.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"d75b3a58fbfe","published":"2025-08-26T21:40:04.000Z","title":"GHSA-q77w-mwjj-7mqx: Picklescan is missing detection when calling built-in python library asyncio.unix_events._UnixSubprocessTransport._start","url":"https://github.com/advisories/GHSA-q77w-mwjj-7mqx","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"dfae661c9135","published":"2025-08-26T21:39:39.000Z","title":"GHSA-49gj-c84q-6qm9: Picklescan is missing detection when calling built-in python cProfile.run","url":"https://github.com/advisories/GHSA-49gj-c84q-6qm9","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"f1d65ba910b7","published":"2025-08-26T21:38:36.000Z","title":"GHSA-9w88-8rmg-7g2p / CVE-2025-71378: Picklescan is missing detection when calling built-in Python cProfile.runctx","url":"https://github.com/advisories/GHSA-9w88-8rmg-7g2p","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b27aeb92493c","published":"2025-08-26T21:38:09.000Z","title":"GHSA-fqq6-7vqf-w3fg: Picklescan is missing detection when calling built-in python doctest.debug_script","url":"https://github.com/advisories/GHSA-fqq6-7vqf-w3fg","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ad90dc96cc9d","published":"2025-08-26T21:37:48.000Z","title":"GHSA-3gf5-cxq9-w223 / CVE-2025-71340: Picklescan is missing detection when calling built-in python idlelib.pyshell.ModifiedInterpreter.runcode","url":"https://github.com/advisories/GHSA-3gf5-cxq9-w223","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"809cea4a6e9c","published":"2025-08-26T21:37:00.000Z","title":"Citrix fixes critical NetScaler RCE flaw exploited in zero-day attacks","url":"https://bleepingcomputer.com/news/security/citrix-fixes-critical-netscaler-rce-flaw-exploited-in-zero-day-attacks","summary":"Citrix fixed three NetScaler ADC and NetScaler Gateway flaws today, including a critical remote code execution flaw tracked as CVE-2025-7775 that was actively exploited in attacks as a zero-day vulnerability.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"incident"},{"id":"a2d339f9f62d","published":"2025-08-26T21:36:47.000Z","title":"GHSA-j343-8v2j-ff7w / CVE-2025-71357: Picklescan is missing detection when calling built-in python idlelib.pyshell.ModifiedInterpreter.runcommand","url":"https://github.com/advisories/GHSA-j343-8v2j-ff7w","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a478a5f7370b","published":"2025-08-26T21:36:45.000Z","title":"On Surjectivity of Neural Networks: Can you elicit any behavior from your model?","url":"https://arxiv.org/abs/2508.19445","summary":"Given a trained neural network, can any specified output be generated by some input? Equivalently, does the network correspond to a function that is surjective? In generative models, surjectivity implies that any output, including harmful or undesirable content, can in…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"c1ce95666765","published":"2025-08-26T21:35:06.000Z","title":"GHSA-m869-42cg-3xwr: Picklescan is missing detection when calling built-in python idlelib.run.Executive.runcode","url":"https://github.com/advisories/GHSA-m869-42cg-3xwr","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c3815d273f5c","published":"2025-08-26T21:34:56.000Z","title":"GHSA-p9w7-82w4-7q8m: Picklescan is missing detection when calling built-in python lib2to3.pgen2.pgen.ParserGenerator.make_label","url":"https://github.com/advisories/GHSA-p9w7-82w4-7q8m","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"31a9fe471f70","published":"2025-08-26T21:34:37.000Z","title":"GHSA-xp4f-hrf8-rxw7 / CVE-2025-71344: Picklescan is missing detection when calling built-in python ensurepip._run_pip","url":"https://github.com/advisories/GHSA-xp4f-hrf8-rxw7","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a77a79aed699","published":"2025-08-26T21:31:07.000Z","title":"GHSA-gqp9-jh35-439m / CVE-2025-52353: Badaso CMS file upload vulnerability","url":"https://github.com/advisories/GHSA-gqp9-jh35-439m","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"71574b4bc264","published":"2025-08-26T21:25:34.000Z","title":"GHSA-4whj-rm5r-c2v8: Picklescan is missing detection when calling pytorch function torch.utils.bottleneck.__main__.run_autograd_prof","url":"https://github.com/advisories/GHSA-4whj-rm5r-c2v8","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"47e88d9a0b46","published":"2025-08-26T20:33:53.000Z","title":"Releases now support immutability in public preview","url":"https://github.blog/changelog/2025-08-26-releases-now-support-immutability-in-public-preview","summary":"Releases now support immutability in public preview GitHub releases now support immutability, adding a new layer of supply chain security. With immutable releases, assets and tags are protected from tampering…","feed":"gh-changelog-supply","publisher":"github","topics":["sca","trust"],"type":"analysis"},{"id":"7c1d5124220e","published":"2025-08-26T20:32:12.000Z","title":"Hundreds of Salesforce customers impacted by attack spree linked to third-party AI agent","url":"https://cyberscoop.com/salesforce-salesloft-drift-attack-spree-google","summary":"A threat group Google tracks as UNC6395 systematically stole large amounts of data from Salesforce customer instances by using OAuth tokens stolen from Salesloft Drift, researchers said.","feed":"cyberscoop","publisher":"cyberscoop","topics":["ai-security"],"type":"research"},{"id":"1497ec8f9299","published":"2025-08-26T20:20:42.000Z","title":"Researchers flag code that uses AI systems to carry out ransomware attacks","url":"https://cyberscoop.com/prompt-lock-eset-ransomware-research-ai-powered-prompt-injection","summary":"The malware, called PromptLock, essentially functions as a hard-coded prompt injection attack on a large language model, inspecting local filesystems, exfiltrating files and encrypting data.","feed":"cyberscoop","publisher":"cyberscoop","topics":["ai-security"],"type":"incident"},{"id":"6299251c1472","published":"2025-08-26T20:18:57.000Z","title":"DOGE employees uploaded Social Security database to ‘vulnerable’ cloud, agency whistleblower says","url":"https://fedscoop.com/doge-social-security-database-whistleblower-cloud-environment-data-vulnerabilities","summary":"","feed":"cyberscoop","publisher":"cyberscoop","topics":[],"type":"analysis"},{"id":"59c0a1a0e6e6","published":"2025-08-26T20:00:59.000Z","title":"AWS Kiro: Arbitrary Code Execution via Indirect Prompt Injection","url":"https://twitter.com/wunderwuzzi23/status/1960365246301995194","summary":"Hacker News: 5 points, 0 comments.","feed":"hn-ai-security-0","publisher":"hacker-news","topics":["ai-security"],"type":"analysis"},{"id":"a99f0cca4928","published":"2025-08-26T19:40:33.000Z","title":"A NIS2 pan-European registry for identifying and classifying essential and important entities","url":"https://arxiv.org/abs/2508.19395","summary":"The NIS2 Directive establishes a common cybersecurity governance model across the European Union, requiring member states to identify, classify, and supervise essential and important entities. As part of a broader governance network, member states are also obligated to notify…","feed":"arxiv-sca","publisher":"arxiv","topics":["trust"],"type":"research"},{"id":"f26ca0818ad0","published":"2025-08-26T18:45:55.000Z","title":"GHSA-224p-v68g-5g8f: GraphQL Armor Max-Depth Plugin Bypass via fragment caching","url":"https://github.com/advisories/GHSA-224p-v68g-5g8f","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ebf5a11db75e","published":"2025-08-26T18:42:37.000Z","title":"GHSA-hmfr-rx46-4jx2: GraphQL Armor Max-Depth Plugin Bypass via Introspection Query Obfuscation","url":"https://github.com/advisories/GHSA-hmfr-rx46-4jx2","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"92dbe7b409f8","published":"2025-08-26T18:39:01.000Z","title":"GHSA-9xph-j2h6-g47v: Picklescan has a missing detection when calling built-in python library idlelib.calltip.get_entity","url":"https://github.com/advisories/GHSA-9xph-j2h6-g47v","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ebfa9a66c36a","published":"2025-08-26T18:38:43.000Z","title":"GHSA-8r4j-24qv-fmq9 / CVE-2025-71361: Picklescan has a missing detection when calling built-in python idlelib.calltip.Calltip","url":"https://github.com/advisories/GHSA-8r4j-24qv-fmq9","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"462a3ebe02c3","published":"2025-08-26T18:37:50.000Z","title":"GHSA-cj3c-v495-4xqh: Picklescan has a missing detection when calling built-in python code.InteractiveInterpreter","url":"https://github.com/advisories/GHSA-cj3c-v495-4xqh","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a8343d9223a4","published":"2025-08-26T18:37:37.000Z","title":"GHSA-7cq8-mj8x-j263 / CVE-2025-71376: Picklescan has a missing detection when calling built-in python idlelib.autocomplete.AutoComplete.fetch_completions","url":"https://github.com/advisories/GHSA-7cq8-mj8x-j263","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"3b9309f54e20","published":"2025-08-26T18:37:10.000Z","title":"GHSA-6w4w-5w54-rjvr / CVE-2025-71358: Picklescan has a missing detection when calling built-in python idlelib.autocomplete.AutoComplete.get_entity","url":"https://github.com/advisories/GHSA-6w4w-5w54-rjvr","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"1d1cb3e657bb","published":"2025-08-26T18:36:42.000Z","title":"GHSA-3vg9-h568-4w9m / CVE-2025-71354: Picklescan has a missing detection when calling built-in python idlelib.debugobj.ObjectTreeItem","url":"https://github.com/advisories/GHSA-3vg9-h568-4w9m","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"08b37108afb6","published":"2025-08-26T18:36:02.000Z","title":"GHSA-f54q-57x4-jg88: Picklescan has a missing detection when calling built-in python lib2to3.pgen2.grammar.Grammar.loads","url":"https://github.com/advisories/GHSA-f54q-57x4-jg88","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"5ebd325ac9e9","published":"2025-08-26T18:35:47.000Z","title":"GHSA-6vqj-c2q5-j97w / CVE-2025-71341: Picklescan has a missing detection when calling built-in python profile.Profile.runctx","url":"https://github.com/advisories/GHSA-6vqj-c2q5-j97w","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"5ddf9da07aa2","published":"2025-08-26T18:35:35.000Z","title":"GHSA-x696-vm39-cp64: Picklescan has a missing detection when calling built-in python profile.Profile.run","url":"https://github.com/advisories/GHSA-x696-vm39-cp64","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"167acf4c2f9c","published":"2025-08-26T18:35:23.000Z","title":"GHSA-g344-hcph-8vgg: Picklescan has a missing detection when calling built-in python trace.Trace.runctx","url":"https://github.com/advisories/GHSA-g344-hcph-8vgg","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"eccdaad67283","published":"2025-08-26T18:35:10.000Z","title":"GHSA-5qwp-399c-mjwf: Picklescan has a missing detection when calling built-in python trace.Trace.run","url":"https://github.com/advisories/GHSA-5qwp-399c-mjwf","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"4753da3424cf","published":"2025-08-26T17:45:11.000Z","title":"GHSA-cfmv-h8fx-85m7 / CVE-2025-11058: xml2rfc has an arbitrary file read vulnerability","url":"https://github.com/advisories/GHSA-cfmv-h8fx-85m7","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ab015599c16d","published":"2025-08-26T17:30:30.000Z","title":"DOGE uploaded live copy of Social Security database to ‘vulnerable’ cloud server, says whistleblower","url":"https://techcrunch.com/2025/08/26/doge-uploaded-live-copy-of-social-security-database-to-vulnerable-cloud-server-says-whistleblower","summary":"The Social Security Administration's chief data officer has publicly blown the whistle, alleging DOGE put hundreds of millions of Social Security records at risk of compromise.","feed":"techcrunch-security","publisher":"techcrunch-security","topics":[],"type":"analysis"},{"id":"656adb2eb8e9","published":"2025-08-26T17:29:00.002Z","title":"Citrix Patches Three NetScaler Flaws, Confirms Active Exploitation of CVE-2025-7775","url":"https://thehackernews.com/2025/08/citrix-patches-three-netscaler-flaws.html","summary":"Citrix has released fixes to address three security flaws in NetScaler ADC and NetScaler Gateway, including one that it said has been actively exploited in the wild. The vulnerabilities in question are listed below - CVE-2025-7775 (CVSS score: 9.2) - Memory overflow…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"advisory"},{"id":"8851ce099e49","published":"2025-08-26T16:19:41.000Z","title":"GHSA-27r7-3m9x-r533 / CVE-2025-57813: traQ Allows Insertion of Sensitive Information into Log File","url":"https://github.com/advisories/GHSA-27r7-3m9x-r533","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"9b5a743d8568","published":"2025-08-26T16:19:04.000Z","title":"GHSA-8mvj-3j78-4qmw / CVE-2025-57810: jsPDF Denial of Service (DoS)","url":"https://github.com/advisories/GHSA-8mvj-3j78-4qmw","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d67fc6f96e10","published":"2025-08-26T16:07:27.000Z","title":"GHSA-mxvv-97wh-cfmm / CVE-2025-57803: ImageMagick (WriteBMPImage): 32-bit integer overflow when writing BMP scanline stride → heap buffer overflow","url":"https://github.com/advisories/GHSA-mxvv-97wh-cfmm","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"fff86c0161f6","published":"2025-08-26T16:00:00.000Z","title":"Securing and governing the rise of autonomous agents","url":"https://microsoft.com/en-us/security/blog/2025/08/26/securing-and-governing-the-rise-of-autonomous-agents","summary":"Hear directly from Corporate Vice President and Deputy Chief Information Security Officer (CISO) for Identity, Igor Sakhnov, about how to secure and govern autonomous agents. This blog is part of a new ongoing series where our Deputy CISOs share their thoughts on what is most…","feed":"microsoft-security","publisher":"microsoft","topics":["trust"],"type":"analysis"},{"id":"61eb1ff8b6a0","published":"2025-08-26T15:29:33.000Z","title":"GHSA-9ccg-6pjw-x645 / CVE-2025-55298: ImageMagick has a Format String Bug in InterpretImageFilename leads to arbitrary code execution","url":"https://github.com/advisories/GHSA-9ccg-6pjw-x645","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6b504a9285b7","published":"2025-08-26T15:27:25.000Z","title":"GHSA-fh55-q5pj-pxgw / CVE-2025-55212: ImageMagick affected by divide-by-zero in ThumbnailImage via montage -geometry \":\" leads to crash","url":"https://github.com/advisories/GHSA-fh55-q5pj-pxgw","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"da2ad5fd621b","published":"2025-08-26T14:30:59.000Z","title":"An Investigation on Group Query Hallucination Attacks","url":"https://arxiv.org/abs/2508.19321","summary":"With the widespread use of large language models (LLMs), understanding their potential failure modes during user interactions is essential. In practice, users often pose multiple questions in a single conversation with LLMs. Therefore, in this study, we propose Group Query…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"82130ed326a9","published":"2025-08-26T14:24:45.000Z","title":"Tackling Federated Unlearning as a Parameter Estimation Problem","url":"https://arxiv.org/abs/2508.19065","summary":"Privacy regulations require the erasure of data from deep learning models. This is a significant challenge that is amplified in Federated Learning, where data remains on clients, making full retraining or coordinated updates often infeasible. This work introduces an efficient…","feed":"arxiv-models","publisher":"arxiv","topics":["trust","models"],"type":"research"},{"id":"c678a8202664","published":"2025-08-26T14:23:00.000Z","title":"Security researcher maps hundreds of TeslaMate servers spilling Tesla vehicle data","url":"https://techcrunch.com/2025/08/26/security-researcher-maps-hundreds-of-teslamate-servers-spilling-tesla-vehicle-data","summary":"A security researcher mapped TeslaMate servers that are publicly exposed to the internet, including sensitive Tesla vehicle data, such as their locations.","feed":"techcrunch-security","publisher":"techcrunch-security","topics":[],"type":"research"},{"id":"f661fa45f828","published":"2025-08-26T14:08:54.000Z","title":"Secrets Found. Owners Identified. Issues Fixed.","url":"https://wiz.io/blog/secrets-found-owners-identified-issues-resolved","summary":"Wiz closes the loop on exposed secrets with blast radius context, ownership intelligence, and actionable, AI-powered fixes.","feed":"wiz","publisher":"wiz","topics":[],"type":"analysis"},{"id":"a691b0087357","published":"2025-08-26T14:00:58.000Z","title":"AWS Kiro: Arbitrary Code Execution via Indirect Prompt Injection","url":"https://embracethered.com/blog/posts/2025/aws-kiro-aribtrary-command-execution-with-indirect-prompt-injection","summary":"On the day AWS Kiro was released, I couldn’t resist putting it through some of my Month of AI Bugs security tests for coding agents. AWS Kiro was vulnerable to arbitrary command execution via indirect prompt injection. This means that a remote attacker, who controls data that…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","ai-security"],"type":"analysis"},{"id":"61b4f0719784","published":"2025-08-26T13:48:00.000Z","title":"Nissan confirms design studio data breach claimed by Qilin ransomware","url":"https://bleepingcomputer.com/news/security/nissan-confirms-design-studio-data-breach-claimed-by-qilin-ransomware","summary":"Nissan Japan has confirmed to BleepingComputer that it suffered a data breach following unauthorized access to a server of one of its subsidiaries, Creative Box Inc. (CBI).","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"7d2acf56e8c4","published":"2025-08-26T13:06:33.000Z","title":"Show HN: 70 Days → 800 GitHub Stars (Cold Start) – My Secret Was a Problem Map","url":"https://github.com/onestardao/WFGY/blob/main/ProblemMap/README.md","summary":"Hacker News: 5 points, 1 comments.","feed":"hn-ai-security-0","publisher":"hacker-news","topics":[],"type":"analysis"},{"id":"9d6559bd3bba","published":"2025-08-26T12:48:05.000Z","title":"GitTaskBench: A Benchmark for Code Agents Solving Real-World Tasks Through Code Repository Leveraging","url":"https://arxiv.org/abs/2508.18993","summary":"Beyond scratch coding, exploiting large-scale code repositories (e.g., GitHub) for practical tasks is vital in real-world software development, yet current benchmarks rarely evaluate code agents in such authentic, workflow-driven scenarios. To bridge this gap, we introduce…","feed":"arxiv-sca","publisher":"arxiv","topics":["agentic-sdlc"],"type":"research"},{"id":"e8b9eafe038a","published":"2025-08-26T11:20:39.000Z","title":"VISION: Robust and Interpretable Code Vulnerability Detection Leveraging Counterfactual Augmentation","url":"https://arxiv.org/abs/2508.18933","summary":"Automated detection of vulnerabilities in source code is an essential cybersecurity challenge, underpinning trust in digital systems and services. Graph Neural Networks (GNNs) have emerged as a promising approach as they can learn structural and logical code relationships in a…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"4e60077dac90","published":"2025-08-26T06:27:00.002Z","title":"Google to Verify All Android Developers in 4 Countries to Block Malicious Apps","url":"https://thehackernews.com/2025/08/google-to-verify-all-android-developers.html","summary":"Google has announced plans to begin verifying the identity of all developers who distribute apps on Android, even for those who distribute their software outside the Play Store. \"Android will require all apps to be registered by verified developers in order to be installed by…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"88ac40241a9b","published":"2025-08-26T06:00:18.000Z","title":"What’s in the SOSS? Podcast #38 – S2E15 Securing AI: A Conversation with Sarah Evans on OpenSSF’s AI/ML Initiatives","url":"https://openssf.org/podcast/2025/08/26/whats-in-the-soss-podcast-38-securing-ai-a-conversation-with-sarah-evans-on-openssfs-ai-ml-initiatives","summary":"Summary In this episode of “What’s in the SOSS,” we welcome back Sarah Evans, Distinguished Engineer at Dell Technologies and a key figure in the OpenSSF’s AI/ML Security working group. Sarah discusses the critical work being done to extend secure software development practices…","feed":"openssf","publisher":"openssf","topics":[],"type":"roundup"},{"id":"4bc624f8c60f","published":"2025-08-26T00:31:14.000Z","title":"GHSA-2f28-69j7-85hf / CVE-2025-50383: Easy!Appointments SQL injection vulnerability","url":"https://github.com/advisories/GHSA-2f28-69j7-85hf","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a0d71702e3bd","published":"2025-08-26T00:31:13.000Z","title":"GHSA-7753-xrfw-ch36 / CVE-2025-5302: LlamaIndex affected by a Denial of Service (DOS) in JSONReader","url":"https://github.com/advisories/GHSA-7753-xrfw-ch36","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"8fe4270e8506","published":"2025-08-26T00:00:00.000Z","title":"CVE-2025-52882: WebSocket authentication bypass in Claude Code extensions","url":"https://securitylabs.datadoghq.com/articles/claude-mcp-cve-2025-52882","summary":"A critical vulnerability in older versions of the Claude Code for Visual Studio Code (VS Code) and other IDE extensions allowed malicious websites to connect to unauthenticated local WebSocket servers, potentially enabling remote command execution","feed":"datadog-seclabs","publisher":"datadog","topics":["sca","agentic-sdlc"],"type":"advisory"},{"id":"d0dfbf3f1aa3","published":"2025-08-26T00:00:00.000Z","title":"CVE-2025-7775: Citrix NetScaler Memory Overflow Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2025-7775","summary":"Citrix NetScaler ADC and NetScaler Gateway contain a memory overflow vulnerability that could allow for remote code execution and/or denial of service. US federal agencies must act by 2025-08-28.","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"4922313a8f3e","published":"2025-08-25T21:48:01.000Z","title":"GHSA-pw25-c82r-75mm / CVE-2025-57814: request-filtering-agent SSRF Bypass via HTTPS Requests to 127.0.0.1","url":"https://github.com/advisories/GHSA-pw25-c82r-75mm","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c2be5323a98a","published":"2025-08-25T21:34:00.000Z","title":"New AI attack hides data-theft prompts in downscaled images","url":"https://bleepingcomputer.com/news/security/new-ai-attack-hides-data-theft-prompts-in-downscaled-images","summary":"Researchers have developed a novel attack that steals user data by injecting malicious prompts in images processed by AI systems before delivering them to a large language model.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"research"},{"id":"c47da86ad3d6","published":"2025-08-25T21:01:00.000Z","title":"GHSA-63cx-g855-hvv4: mitmproxy binaries embed a vulnerable python-hyper/h2 dependency","url":"https://github.com/advisories/GHSA-63cx-g855-hvv4","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"494bef273d87","published":"2025-08-25T20:44:43.000Z","title":"GHSA-847f-9342-265h / CVE-2025-57804: h2 allows HTTP Request Smuggling due to illegal characters in headers","url":"https://github.com/advisories/GHSA-847f-9342-265h","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca","dast"],"type":"advisory"},{"id":"f617878b44fa","published":"2025-08-25T20:43:45.000Z","title":"GHSA-5cmr-4px5-23pc / CVE-2025-57809: XGrammar affected by Denial of Service by infinite recursion grammars","url":"https://github.com/advisories/GHSA-5cmr-4px5-23pc","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6252d982161d","published":"2025-08-25T20:42:45.000Z","title":"GHSA-crcq-738g-pqvc / CVE-2025-57811: Craft CMS Potential Remote Code Execution via Twig SSTI","url":"https://github.com/advisories/GHSA-crcq-738g-pqvc","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"1300cd5bee92","published":"2025-08-25T17:53:00.000Z","title":"Docker Fixes CVE-2025-9074, Critical Container Escape Vulnerability With CVSS Score 9.3","url":"https://thehackernews.com/2025/08/docker-fixes-cve-2025-9074-critical.html","summary":"Docker has released fixes to address a critical security flaw affecting the Docker Desktop app for Windows and macOS that could potentially allow an attacker to break out of the confines of a container. The vulnerability, tracked as CVE-2025-9074, carries a CVSS score of 9.3…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"analysis"},{"id":"445ae25eeba6","published":"2025-08-25T16:37:00.000Z","title":"Malicious Android apps with 19M installs removed from Google Play","url":"https://bleepingcomputer.com/news/security/malicious-android-apps-with-19m-installs-removed-from-google-play","summary":"Seventy-seven malicious Android apps containing different types of malware were found on Google Play after being downloaded more than 19 million times.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"f9a4ee2716f2","published":"2025-08-25T16:23:51.000Z","title":"GHSA-6hgw-6x87-578x / CVE-2025-55160: ImageMagick has Undefined Behavior (function-type-mismatch) in CloneSplayTree","url":"https://github.com/advisories/GHSA-6hgw-6x87-578x","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b18cbef5aab6","published":"2025-08-25T16:23:09.000Z","title":"GHSA-qp29-wxp5-wh82 / CVE-2025-55154: imagemagick: integer overflows in MNG magnification","url":"https://github.com/advisories/GHSA-qp29-wxp5-wh82","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"597ac142746b","published":"2025-08-25T16:21:04.000Z","title":"GHSA-4gv9-mp8m-592r / CVE-2025-57760: Langflow Vulnerable to Privilege Escalation via CLI Superuser Creation (Post-RCE)","url":"https://github.com/advisories/GHSA-4gv9-mp8m-592r","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"61ae5a5b2619","published":"2025-08-25T16:01:12.000Z","title":"Safeguarding VS Code against prompt injections","url":"https://github.blog/security/vulnerability-research/safeguarding-vs-code-against-prompt-injections","summary":"When a chat conversation is poisoned by indirect prompt injection, it can result in the exposure of GitHub tokens, confidential files, or even the execution of arbitrary code without the user's explicit consent. In this blog post, we'll explain which VS Code features may reduce…","feed":"github-security","publisher":"github","topics":["agentic-sdlc","ai-security"],"type":"analysis"},{"id":"4a782c630e20","published":"2025-08-25T15:58:40.000Z","title":"GHSA-cjc8-g9w8-chfw / CVE-2025-55004: imagemagick: heap-buffer overflow read in MNG magnification with alpha","url":"https://github.com/advisories/GHSA-cjc8-g9w8-chfw","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"25d002094197","published":"2025-08-25T15:54:51.000Z","title":"GHSA-fff3-4rp7-px97 / CVE-2025-68469: ImageMagick has a heap-buffer-overflow","url":"https://github.com/advisories/GHSA-fff3-4rp7-px97","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"400ab2932590","published":"2025-08-25T15:53:57.000Z","title":"GHSA-cfh4-9f7v-fhrc / CVE-2025-53019: ImageMagick has a Memory Leak in magick stream","url":"https://github.com/advisories/GHSA-cfh4-9f7v-fhrc","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d03e41a496de","published":"2025-08-25T15:53:34.000Z","title":"GHSA-hm4x-r5hc-794f / CVE-2025-53014: ImageMagick has a Heap Buffer Overflow in InterpretImageFilename","url":"https://github.com/advisories/GHSA-hm4x-r5hc-794f","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"fddf915496fd","published":"2025-08-25T15:43:37.000Z","title":"GHSA-qh3h-j545-h8c9 / CVE-2025-53101: ImageMagick has a Stack Buffer Overflow in image.c","url":"https://github.com/advisories/GHSA-qh3h-j545-h8c9","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"0dd6acc0c882","published":"2025-08-25T15:32:00.000Z","title":"GHSA-5c4f-pxmx-xcm4 / CVE-2025-26467: Apache Cassandra: User with MODIFY permission on ALL KEYSPACES can escalate privileges to superuser via unsafe actions (4.0.16 only)","url":"https://github.com/advisories/GHSA-5c4f-pxmx-xcm4","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"8346e5081636","published":"2025-08-25T15:32:00.000Z","title":"GHSA-mqh4-2mm8-g7w9 / CVE-2025-43960: Adminer PHP Object Injection issue leads to Denial of Service","url":"https://github.com/advisories/GHSA-mqh4-2mm8-g7w9","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"3e5ba16b4d62","published":"2025-08-25T15:11:11.000Z","title":"A.S.E: A Repository-Level Benchmark for Evaluating Security in AI-Generated Code","url":"https://arxiv.org/abs/2508.18106","summary":"The increasing adoption of large language models (LLMs) in software engineering necessitates rigorous security evaluation of their generated code. However, existing benchmarks often lack relevance to real-world AI-assisted programming scenarios, making them inadequate for…","feed":"arxiv-agentic","publisher":"arxiv","topics":["agentic-sdlc"],"type":"research"},{"id":"d044e2990613","published":"2025-08-25T15:11:00.000Z","title":"Critical Docker Desktop flaw lets attackers hijack Windows hosts","url":"https://bleepingcomputer.com/news/security/critical-docker-desktop-flaw-lets-attackers-hijack-windows-hosts","summary":"A critical vulnerability in Docker Desktop for Windows and macOS allows compromising the host by running a malicious container, even if the Enhanced Container Isolation (ECI) protection is active.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"advisory"},{"id":"279b78b6bdb5","published":"2025-08-25T14:32:32.000Z","title":"GHSA-rx7m-68vc-ppxh / CVE-2025-54370: PhpSpreadsheet vulnerable to SSRF when reading and displaying a processed HTML document in the browser","url":"https://github.com/advisories/GHSA-rx7m-68vc-ppxh","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a0ed9773ba28","published":"2025-08-25T14:16:30.000Z","title":"Stand on The Shoulders of Giants: Building JailExpert from Previous Attack Experience","url":"https://arxiv.org/abs/2508.19292","summary":"Large language models (LLMs) generate human-aligned content under certain safety constraints. However, the current known technique ``jailbreak prompt'' can circumvent safety-aligned measures and induce LLMs to output malicious content. Research on Jailbreaking can help identify…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"09c649739e7e","published":"2025-08-25T12:25:28.000Z","title":"MoveScanner: Analysis of Security Risks of Move Smart Contracts","url":"https://arxiv.org/abs/2508.17964","summary":"As blockchain technology continues to evolve, the security of smart contracts has increasingly drawn attention from both academia and industry. The Move language, with its unique resource model and linear type system, provides a solid foundation for the security of digital…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"558c18905e81","published":"2025-08-25T12:17:00.005Z","title":"⚡ Weekly Recap: Password Manager Flaws, Apple 0-Day, Hidden AI Prompts, In-the-Wild Exploits & More","url":"https://thehackernews.com/2025/08/weekly-recap-password-manager-flaws.html","summary":"Cybersecurity today moves at the pace of global politics. A single breach can ripple across supply chains, turn a software flaw into leverage, or shift who holds the upper hand. For leaders, this means defense isn’t just a matter of firewalls and patches—it’s about strategy.…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"roundup"},{"id":"047e6590a7db","published":"2025-08-25T11:00:58.000Z","title":"How Prompt Injection Exposes Manus' VS Code Server to the Internet","url":"https://embracethered.com/blog/posts/2025/manus-ai-kill-chain-expose-port-vs-code-server-on-internet","summary":"Today we will cover a powerful, easy to use, autonomous agent called Manus. Manus is developed by the Chinese startup Butterfly Effect , headquartered in Singapore. This post demonstrates an end-to-end indirect prompt injection attack leading to a compromise of Manus’ dev box.…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["ai-security"],"type":"analysis"},{"id":"116821c12b51","published":"2025-08-25T10:45:10.000Z","title":"PhantomLint: Principled Detection of Hidden LLM Prompts in Structured Documents","url":"https://arxiv.org/abs/2508.17884","summary":"Hidden LLM prompts have appeared in online documents with increasing frequency. Their goal is to trigger indirect prompt injection attacks while remaining undetected from human oversight, to manipulate LLM-powered automated document processing systems, against applications as…","feed":"arxiv-aisec","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"13fdb86b50b8","published":"2025-08-25T09:39:15.000Z","title":"Agentic Browser Security: Indirect Prompt Injection in Perplexity Comet","url":"https://simonwillison.net/2025/Aug/25/agentic-browser-security","summary":"Agentic Browser Security: Indirect Prompt Injection in Perplexity Comet The security team from Brave took a look at Comet, the LLM-powered \"agentic browser\" extension from Perplexity, and unsurprisingly found security holes you can drive a truck through. The vulnerability we’re…","feed":"simonw","publisher":"simonw","topics":["ai-security"],"type":"analysis"},{"id":"8a39d163b1e6","published":"2025-08-25T07:30:10.000Z","title":"Speculative Safety-Aware Decoding","url":"https://arxiv.org/abs/2508.17739","summary":"Despite extensive efforts to align Large Language Models (LLMs) with human values and safety rules, jailbreak attacks that exploit certain vulnerabilities continuously emerge, highlighting the need to strengthen existing LLMs with additional safety properties to defend against…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"909a680795af","published":"2025-08-25T05:25:28.000Z","title":"Tricking LLM-Based NPCs into Spilling Secrets","url":"https://arxiv.org/abs/2508.19288","summary":"Large Language Models (LLMs) are increasingly used to generate dynamic dialogue for game NPCs. However, their integration raises new security concerns. In this study, we examine whether adversarial prompt injection can cause LLM-based NPCs to reveal hidden background secrets…","feed":"arxiv-aisec","publisher":"arxiv","topics":["models","ai-security"],"type":"research"},{"id":"580b5c3884ee","published":"2025-08-25T05:13:23.000Z","title":"Attacking LLMs and AI Agents: Advertisement Embedding Attacks Against Large Language Models","url":"https://arxiv.org/abs/2508.17674","summary":"We introduce Advertisement Embedding Attacks (AEA), a new class of LLM security threats that stealthily inject promotional or malicious content into model outputs and AI agents. AEA operate through two low-cost vectors: (1) hijacking third-party service-distribution platforms…","feed":"arxiv-agentic","publisher":"arxiv","topics":["models","ai-security"],"type":"research"},{"id":"981e6d6ed723","published":"2025-08-25T00:00:00.000Z","title":"CVE-2024-8068: Citrix Session Recording Improper Privilege Management Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2024-8068","summary":"Citrix Session Recording contains an improper privilege management vulnerability that could allow for privilege escalation to NetworkService Account access. An attacker must be an authenticated user in the same Windows Active Directory domain as the session recording server domai","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"a5a744345364","published":"2025-08-25T00:00:00.000Z","title":"CVE-2024-8069: Citrix Session Recording Deserialization of Untrusted Data Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2024-8069","summary":"Citrix Session Recording contains a deserialization of untrusted data vulnerability that allows limited remote code execution with privilege of a NetworkService Account access. Attacker must be an authenticated user on the same intranet as the session recording server. US federal","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"fc7f32d3f368","published":"2025-08-25T00:00:00.000Z","title":"CVE-2025-48384: Git Link Following Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2025-48384","summary":"Git contains a link following vulnerability that stems from Git’s inconsistent handling of carriage return characters in configuration files. US federal agencies must act by 2025-09-15.","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"92e84fbd625f","published":"2025-08-24T22:22:09.000Z","title":"Activation Transport Operators","url":"https://arxiv.org/abs/2508.17540","summary":"The residual stream mediates communication between transformer decoder layers via linear reads and writes of non-linear computations. While sparse-dictionary learning-based methods locate features in the residual stream, and activation patching methods discover circuits within…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"68a5f8b869c3","published":"2025-08-24T14:08:00.000Z","title":"New Android malware poses as antivirus from Russian intelligence agency","url":"https://bleepingcomputer.com/news/security/new-android-malware-poses-as-antivirus-from-russian-intelligence-agency","summary":"A new Android malware posing as an antivirus tool software created by Russia's Federal Security Services agency (FSB) is being used to target executives of Russian businesses.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"5a9a2268f3f1","published":"2025-08-24T13:42:48.000Z","title":"Trust Me, I Know This Function: Hijacking LLM Static Analysis using Bias","url":"https://arxiv.org/abs/2508.17361","summary":"Large Language Models (LLMs) are increasingly trusted to perform automated code review and static analysis at scale, supporting tasks such as vulnerability detection, summarization, and refactoring. In this paper, we identify and exploit a critical vulnerability in LLM-based…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"0b226f0d55b9","published":"2025-08-24T13:38:00.003Z","title":"Malicious Go Module Poses as SSH Brute-Force Tool, Steals Credentials via Telegram Bot","url":"https://thehackernews.com/2025/08/malicious-go-module-poses-as-ssh-brute.html","summary":"Cybersecurity researchers have discovered a malicious Go module that presents itself as a brute-force tool for SSH but actually contains functionality to discreetly exfiltrate credentials to its creator. \"On the first successful login, the package sends the target IP address,…","feed":"thehackernews","publisher":"thehackernews","topics":["sca"],"type":"research"},{"id":"754d14b04cbc","published":"2025-08-24T02:52:08.000Z","title":"Agentic Browser Security: Indirect Prompt Injection in Perplexity Comet","url":"https://brave.com/blog/comet-prompt-injection","summary":"Hacker News: 97 points, 31 comments.","feed":"hn-ai-security-0","publisher":"hacker-news","topics":["ai-security"],"type":"analysis"},{"id":"1fa81ca33825","published":"2025-08-23T23:20:58.000Z","title":"Sneaking Invisible Instructions by Developers in Windsurf","url":"https://embracethered.com/blog/posts/2025/windsurf-sneaking-invisible-instructions-for-prompt-injection","summary":"Imagine a malicious instruction hidden in plain sight, invisible to you but not to the AI. This is a vulnerability discovered in Windsurf Cascade, it follows invisible instructions. This means there can be instructions in a file or result of a tool call that the developer…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","ai-security"],"type":"analysis"},{"id":"cb98a285fe50","published":"2025-08-23T23:17:36.000Z","title":"Data exfiltration by AI prompt injection","url":"https://twitter.com/zack_overflow/status/1959308058200551721","summary":"Hacker News: 5 points, 1 comments.","feed":"hn-ai-security-0","publisher":"hacker-news","topics":["ai-security"],"type":"analysis"},{"id":"7c4c31f359a2","published":"2025-08-23T22:41:49.000Z","title":"Mind the Gap: Time-of-Check to Time-of-Use Vulnerabilities in LLM-Enabled Agents","url":"https://arxiv.org/abs/2508.17155","summary":"Large Language Model (LLM)-enabled agents are rapidly emerging across a wide range of applications, but their deployment introduces vulnerabilities with security implications. While prior work has examined prompt-based attacks (e.g., prompt injection) and data-oriented threats…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"e2568c247312","published":"2025-08-23T16:27:42.000Z","title":"POT: Inducing Overthinking in LLMs via Black-Box Iterative Optimization","url":"https://arxiv.org/abs/2508.19277","summary":"Recent advances in Chain-of-Thought (CoT) prompting have substantially enhanced the reasoning capabilities of large language models (LLMs), enabling sophisticated problem-solving through explicit multi-step reasoning traces. However, these enhanced reasoning processes introduce…","feed":"arxiv-sca","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"07fd5bf8a5bb","published":"2025-08-23T07:38:00.002Z","title":"GeoServer Exploits, PolarEdge, and Gayfemboy Push Cybercrime Beyond Traditional Botnets","url":"https://thehackernews.com/2025/08/geoserver-exploits-polaredge-and.html","summary":"Cybersecurity researchers are calling attention to multiple campaigns that are taking advantage of known security vulnerabilities and exposed Redis servers to various malicious activities, including leveraging the compromised devices as IoT botnets, residential proxies, or…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"research"},{"id":"acb6908b1aa0","published":"2025-08-23T06:30:20.000Z","title":"GHSA-h8gx-4hhm-w45v / CVE-2025-43765: Liferay Portal stored cross-site scripting in text field of the web content structure","url":"https://github.com/advisories/GHSA-h8gx-4hhm-w45v","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"e934d28978e8","published":"2025-08-23T06:30:20.000Z","title":"GHSA-mf9q-87xx-jgvv / CVE-2025-43766: Liferay Portal allows unrestricted upload of file in the style books component","url":"https://github.com/advisories/GHSA-mf9q-87xx-jgvv","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"98914f438dbf","published":"2025-08-23T06:30:19.000Z","title":"GHSA-6hj4-v2qp-cqr2 / CVE-2025-43767: Liferay Portal allows open redirect in /c/portal/edit_info_item parameter redirect","url":"https://github.com/advisories/GHSA-6hj4-v2qp-cqr2","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a332812140b6","published":"2025-08-23T06:30:19.000Z","title":"GHSA-23w4-rpc6-wpcc / CVE-2025-43764: Liferay Portal ReDoS with Role Name search in KaleoDesignerPortlet","url":"https://github.com/advisories/GHSA-23w4-rpc6-wpcc","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"0e6090451fab","published":"2025-08-23T03:30:43.000Z","title":"GHSA-rvmf-jw8g-r35r / CVE-2025-43769: Liferay Portal vulnerable to Stored XSS in Components portlet","url":"https://github.com/advisories/GHSA-rvmf-jw8g-r35r","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"94c2fd7be130","published":"2025-08-23T03:30:43.000Z","title":"GHSA-cv9j-mg9w-v7wm / CVE-2025-43768: Liferay Portal JSONWS API endpoint shares sensitive information","url":"https://github.com/advisories/GHSA-cv9j-mg9w-v7wm","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b713dbf6695a","published":"2025-08-23T03:30:43.000Z","title":"GHSA-h4m4-xp33-37mj / CVE-2025-43770: Liferay Portal vulnerable to Reflected XSS with the referer and forward parameter","url":"https://github.com/advisories/GHSA-h4m4-xp33-37mj","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ac2a122eda98","published":"2025-08-22T23:57:04.000Z","title":"A Survey of Threats Against Voice Authentication and Anti-Spoofing Systems","url":"https://arxiv.org/abs/2508.16843","summary":"Voice authentication has undergone significant changes from traditional systems that relied on handcrafted acoustic features to deep learning models that can extract robust speaker embeddings. This advancement has expanded its applications across finance, smart devices, law…","feed":"arxiv-models","publisher":"arxiv","topics":["trust","models"],"type":"research"},{"id":"056ecdcca59f","published":"2025-08-22T21:31:16.000Z","title":"GHSA-w3cr-3xw2-rp78 / CVE-2025-43759: Liferay Portal users are able to add system admin portlets to pages","url":"https://github.com/advisories/GHSA-w3cr-3xw2-rp78","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"49b317ebdf0a","published":"2025-08-22T21:31:16.000Z","title":"GHSA-3h7r-4xxj-3mfm / CVE-2025-43761: Liferay Portal Reflected XSS in CKeditor 4.21.0 endpoint","url":"https://github.com/advisories/GHSA-3h7r-4xxj-3mfm","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"89e96287f1f7","published":"2025-08-22T21:31:16.000Z","title":"GHSA-mm62-gwj5-j285 / CVE-2025-43758: Liferay Portal's unauthenticated users can access loaded files via URL before submitting the object entry","url":"https://github.com/advisories/GHSA-mm62-gwj5-j285","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a09ecbee787f","published":"2025-08-22T21:31:16.000Z","title":"GHSA-84pp-qr92-95c9 / CVE-2025-43762: Liferay Portal users can upload an unlimited amount of files","url":"https://github.com/advisories/GHSA-84pp-qr92-95c9","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"2c11c1f75c22","published":"2025-08-22T20:58:21.000Z","title":"GHSA-95v9-hv42-pwrj / CVE-2025-57801: gnark is vulnerable to signature malleability in EdDSA and ECDSA due to missing scalar checks","url":"https://github.com/advisories/GHSA-95v9-hv42-pwrj","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"aad7166baf9b","published":"2025-08-22T18:31:24.000Z","title":"GHSA-fvqv-593q-qp8r / CVE-2025-43760: Liferay Portal Reflected Cross-Site Scripting Vulnerability via PortalUtil.escapeRedirect","url":"https://github.com/advisories/GHSA-fvqv-593q-qp8r","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6172b67ade95","published":"2025-08-22T18:31:23.000Z","title":"GHSA-xwc5-q44v-p6gg / CVE-2025-43751: Liferay Portal User Enumeration Vulnerability via the Create Account Page","url":"https://github.com/advisories/GHSA-xwc5-q44v-p6gg","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"8eb723323846","published":"2025-08-22T17:34:45.000Z","title":"GHSA-655h-hg88-5qmf: Rust XCB `xcb::Connection::connect_to_fd*` functions violate I/O safety","url":"https://github.com/advisories/GHSA-655h-hg88-5qmf","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"7cb8da2cf641","published":"2025-08-22T16:58:14.000Z","title":"GHSA-vv6j-3g6g-2pvj / CVE-2025-71348: Picklescan is missing detection when calling pytorch function torch.utils._config_module.load_config","url":"https://github.com/advisories/GHSA-vv6j-3g6g-2pvj","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"f466be87246d","published":"2025-08-22T16:58:06.000Z","title":"GHSA-vr7h-p6mm-wpmh / CVE-2025-71370: Picklescan missing detection when calling pytorch function torch.jit.unsupported_tensor_ops.execWrapper","url":"https://github.com/advisories/GHSA-vr7h-p6mm-wpmh","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a0639041d7ab","published":"2025-08-22T16:57:59.000Z","title":"GHSA-h3qp-7fh3-f8h4: Picklescan missing detection when calling pytorch function torch.utils.data.datapipes.utils.decoder.basichandlers","url":"https://github.com/advisories/GHSA-h3qp-7fh3-f8h4","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"71b46ef58149","published":"2025-08-22T16:57:49.000Z","title":"GHSA-f745-w6jp-hpxx: Picklescan missing detection when calling pytorch function torch.utils.collect_env.run","url":"https://github.com/advisories/GHSA-f745-w6jp-hpxx","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"5cdf6173eef0","published":"2025-08-22T16:56:53.000Z","title":"GHSA-f4x7-rfwp-v3xw: Picklescan missing detection when calling pytorch function torch.fx.experimental.symbolic_shapes.ShapeEnv.evaluate_guards_expression","url":"https://github.com/advisories/GHSA-f4x7-rfwp-v3xw","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c5ba0e46a5ed","published":"2025-08-22T16:56:39.000Z","title":"GHSA-86cj-95qr-2p4f: Picklescan missing detection when calling pytorch function torch._dynamo.guards.GuardBuilder.get","url":"https://github.com/advisories/GHSA-86cj-95qr-2p4f","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"16ed9f484a72","published":"2025-08-22T16:56:25.000Z","title":"GHSA-4r9r-ch6f-vxmx: Picklescan missing detection when calling pytorch function torch.utils.bottleneck.__main__.run_cprofile","url":"https://github.com/advisories/GHSA-4r9r-ch6f-vxmx","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"902ba7882859","published":"2025-08-22T16:50:59.000Z","title":"Towards Log Analysis with AI Agents: Cowrie Case Study","url":"https://arxiv.org/abs/2509.05306","summary":"The scarcity of real-world attack data significantly hinders progress in cybersecurity research and education. Although honeypots like Cowrie effectively collect live threat intelligence, they generate overwhelming volumes of unstructured and heterogeneous logs, rendering…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"679861045b6e","published":"2025-08-22T16:50:05.000Z","title":"GHSA-74rg-6f92-g6wx / CVE-2025-55745: UnoPim has CSV Injection on Quick Export feature","url":"https://github.com/advisories/GHSA-74rg-6f92-g6wx","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c8c5d87716c7","published":"2025-08-22T16:49:55.000Z","title":"GHSA-8p2f-fx4q-75cx / CVE-2025-55741: UnoPim has Broken Access Control","url":"https://github.com/advisories/GHSA-8p2f-fx4q-75cx","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ef2a22165cab","published":"2025-08-22T16:49:05.000Z","title":"GHSA-gcqf-pxgg-gw8q / CVE-2025-53363: Dpanel has an arbitrary file read vulnerability","url":"https://github.com/advisories/GHSA-gcqf-pxgg-gw8q","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"fd8b47169482","published":"2025-08-22T15:33:05.000Z","title":"GHSA-gj8w-ffq9-6828 / CVE-2025-51825: JeecgBoot SQL Injection Vulnerability","url":"https://github.com/advisories/GHSA-gj8w-ffq9-6828","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"7d7aee23aff2","published":"2025-08-22T15:17:44.000Z","title":"CelloAI: Leveraging Large Language Models for HPC Software Development in High Energy Physics","url":"https://arxiv.org/abs/2508.16713","summary":"Next-generation High Energy Physics (HEP) experiments will generate unprecedented data volumes, necessitating High Performance Computing (HPC) integration alongside traditional high-throughput computing. However, HPC adoption in HEP is hindered by the challenge of porting…","feed":"arxiv-sca","publisher":"arxiv","topics":["agentic-sdlc"],"type":"research"},{"id":"817faba8e6d4","published":"2025-08-22T15:11:16.000Z","title":"What Not to Miss at Open Source Summit & OpenSSF Community Day Europe","url":"https://openssf.org/blog/2025/08/22/what-not-to-miss-at-open-source-summit-openssf-community-day-europe","summary":"The countdown is on! From August 25 to 28, 2025, the open source security community will gather in Amsterdam for Open Source Summit Europe and OpenSSF Community Day Europe. These two major gatherings will focus on the future of software supply chain security, regulatory…","feed":"openssf","publisher":"openssf","topics":["sca"],"type":"analysis"},{"id":"bb61f216687b","published":"2025-08-22T14:31:00.001Z","title":"Linux Malware Delivered via Malicious RAR Filenames Evades Antivirus Detection","url":"https://thehackernews.com/2025/08/linux-malware-delivered-via-malicious.html","summary":"Cybersecurity researchers have shed light on a novel attack chain that employs phishing emails to deliver an open-source backdoor called VShell. The \"Linux-specific malware infection chain that starts with a spam email with a malicious RAR archive file,\" Trellix researcher…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"research"},{"id":"d215ac63f4a4","published":"2025-08-22T14:26:34.000Z","title":"Developer gets prison time for sabotaging former employer’s network with a ‘kill switch’","url":"https://techcrunch.com/2025/08/22/developer-gets-prison-time-for-sabotaging-former-employers-network-with-a-kill-switch","summary":"The ex-developer was convicted of planting malicious code designed to crash its servers in the event that he was fired.","feed":"techcrunch-security","publisher":"techcrunch-security","topics":[],"type":"analysis"},{"id":"779e51af6c2d","published":"2025-08-22T14:13:16.000Z","title":"Retrieval-Augmented Defense: Adaptive and Controllable Jailbreak Prevention for Large Language Models","url":"https://arxiv.org/abs/2508.16406","summary":"Large Language Models (LLMs) remain vulnerable to jailbreak attacks, which attempt to elicit harmful responses from LLMs. The evolving nature and diversity of these attacks pose many challenges for defense systems, including (1) adaptation to counter emerging attack strategies…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"45ae8e5dfc21","published":"2025-08-22T12:41:26.000Z","title":"Confusion is the Final Barrier: Rethinking Jailbreak Evaluation and Investigating the Real Misuse Threat of LLMs","url":"https://arxiv.org/abs/2508.16347","summary":"With the development of Large Language Models (LLMs), numerous efforts have revealed their vulnerabilities to jailbreak attacks. Although these studies have driven the progress in LLMs' safety alignment, it remains unclear whether LLMs have internalized authentic knowledge to…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"1a38dc9a5853","published":"2025-08-22T12:30:30.000Z","title":"GHSA-g6rx-6wfx-gj74 / CVE-2025-9340: Bouncy Castle for Java has Out-of-Bounds Write Vulnerability","url":"https://github.com/advisories/GHSA-g6rx-6wfx-gj74","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"bd13904ea8a7","published":"2025-08-22T12:25:00.000Z","title":"Microsoft: August Windows updates cause severe streaming issues","url":"https://bleepingcomputer.com/news/microsoft/microsoft-august-windows-updates-cause-severe-ndi-streaming-issues","summary":"Microsoft has confirmed that the August 2025 security updates are causing severe lag and stuttering issues with NDI streaming software on some Windows 10 and Windows 11 systems.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"analysis"},{"id":"2cc393544a79","published":"2025-08-22T11:06:00.000Z","title":"Automation Is Redefining Pentest Delivery","url":"https://thehackernews.com/2025/08/automation-is-redefining-pentest.html","summary":"Pentesting remains one of the most effective ways to identify real-world security weaknesses before adversaries do. But as the threat landscape has evolved, the way we deliver pentest results hasn't kept pace. Most organizations still rely on traditional reporting…","feed":"thehackernews","publisher":"thehackernews","topics":["dast"],"type":"analysis"},{"id":"28e10056f46f","published":"2025-08-22T10:00:00.000Z","title":"Evolving OAuth Client Registration in the Model Context Protocol","url":"https://blog.modelcontextprotocol.io/posts/client_registration","summary":"The Model Context Protocol (MCP) has adopted OAuth 2.1 as the foundation for its authorization framework. A key part of the authorization flow that MCP is particularly reliant on is client registration . This is especially important in a world where clients and servers don’t…","feed":"mcp-blog","publisher":"mcp","topics":["ai-security"],"type":"analysis"},{"id":"c97e5a80634a","published":"2025-08-22T09:30:41.000Z","title":"GHSA-jfcv-jv9g-2vx2 / CVE-2025-9341: Bouncy Castle for Java has Uncontrolled Resource Consumption Vulnerability","url":"https://github.com/advisories/GHSA-jfcv-jv9g-2vx2","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"14dd6b5f21ab","published":"2025-08-22T06:18:57.000Z","title":"The Aegis Protocol: A Foundational Security Framework for Autonomous AI Agents","url":"https://arxiv.org/abs/2508.19267","summary":"The proliferation of autonomous AI agents marks a paradigm shift toward complex, emergent multi-agent systems. This transition introduces systemic security risks, including control-flow hijacking and cascading failures, that traditional cybersecurity paradigms are ill-equipped…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"c5ea389e0a97","published":"2025-08-22T05:23:22.000Z","title":"GHSA-mx4r-mrq4-m6cw: Malware in apollo-ms","url":"https://github.com/advisories/GHSA-mx4r-mrq4-m6cw","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-malware","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"82399f640cf1","published":"2025-08-22T03:30:24.000Z","title":"GHSA-qpp6-f3qj-rggq / CVE-2025-43752: Liferay Portal's Unlimited File Upload Could Result in DoS","url":"https://github.com/advisories/GHSA-qpp6-f3qj-rggq","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"0ead1e165cfe","published":"2025-08-22T00:30:31.000Z","title":"GHSA-r367-q549-pgr5 / CVE-2025-43753: Liferay Portal Reflected Cross-Site Scripting Vulnerability via Form Container","url":"https://github.com/advisories/GHSA-r367-q549-pgr5","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b985953d994e","published":"2025-08-21T23:46:00.000Z","title":"Dev gets 4 years for creating kill switch on ex-employer's systems","url":"https://bleepingcomputer.com/news/security/dev-gets-4-years-for-creating-kill-switch-on-ex-employers-systems","summary":"A software developer has been sentenced to four years in prison for sabotaging his ex-employer's Windows network with custom malware and a kill switch that locked out employees when his account was disabled.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"3a4a0401ffce","published":"2025-08-21T21:32:08.000Z","title":"GHSA-48cg-9c55-j2q7 / CVE-2025-51606: hippo4j Includes Hard Coded Secret Key in JWT Creation","url":"https://github.com/advisories/GHSA-48cg-9c55-j2q7","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"933b68650d0a","published":"2025-08-21T20:11:52.000Z","title":"GHSA-9gjj-6gj7-c4wj / CVE-2025-57751: Denial-of-Service attack in pyLoad CNL Blueprint using dukpy.evaljs","url":"https://github.com/advisories/GHSA-9gjj-6gj7-c4wj","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a6cd0861f74f","published":"2025-08-21T19:53:26.000Z","title":"Strategic Sample Selection for Improved Clean-Label Backdoor Attacks in Text Classification","url":"https://arxiv.org/abs/2508.15934","summary":"Backdoor attacks pose a significant threat to the integrity of text classification models used in natural language processing. While several dirty-label attacks that achieve high attack success rates (ASR) have been proposed, clean-label attacks are inherently more difficult.…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"762a8178b1cc","published":"2025-08-21T18:31:29.000Z","title":"GHSA-h469-4fcf-p23h / CVE-2025-8402: Mattermost has Potential Server Crash due to Unvalidated Import Data","url":"https://github.com/advisories/GHSA-h469-4fcf-p23h","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"aa3d0c5ddfc3","published":"2025-08-21T18:31:29.000Z","title":"GHSA-pj6f-rc94-gw53 / CVE-2025-6465: Mattermost Fails to Sanitize File Names","url":"https://github.com/advisories/GHSA-pj6f-rc94-gw53","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d4ee3c80b75d","published":"2025-08-21T18:31:29.000Z","title":"GHSA-x7p4-v8mj-6fxx / CVE-2025-43754: Liferay Portal Username Enumeration Vulnerability","url":"https://github.com/advisories/GHSA-x7p4-v8mj-6fxx","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a9c6d23d3713","published":"2025-08-21T18:31:28.000Z","title":"GHSA-58cq-8wm2-6m87 / CVE-2025-43755: Liferay Portal Stored Cross-Site Scripting Vulnerability via GroupPagesPortlet_type Parameter","url":"https://github.com/advisories/GHSA-58cq-8wm2-6m87","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d3a70a9a5a52","published":"2025-08-21T18:31:28.000Z","title":"GHSA-q2gv-w583-f2vq / CVE-2025-43756: Liferay Portal Reflected Cross-Site Scripting Vulnerability via snippet Parameter","url":"https://github.com/advisories/GHSA-q2gv-w583-f2vq","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"5b9f6b4b9584","published":"2025-08-21T18:19:06.000Z","title":"OpenSSF Newsletter – August 2025","url":"https://openssf.org/newsletter/2025/08/21/openssf-newsletter-august-2025","summary":"Welcome to the August 2025 edition of the OpenSSF Newsletter! Here’s a roundup of the latest developments, key events, and upcoming opportunities in the Open Source Security community. TL;DR: 🎉...","feed":"openssf","publisher":"openssf","topics":[],"type":"roundup"},{"id":"793da97c3df0","published":"2025-08-21T16:38:00.001Z","title":"Pre-Auth Exploit Chains Found in Commvault Could Enable Remote Code Execution Attacks","url":"https://thehackernews.com/2025/08/pre-auth-exploit-chains-found-in.html","summary":"Commvault has released updates to address four security gaps that could be exploited to achieve remote code execution on susceptible instances. The list of vulnerabilities, identified in Commvault versions before 11.36.60, is as follows - CVE-2025-57788 (CVSS score: 6.9) - A…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"analysis"},{"id":"2c5e93363c22","published":"2025-08-21T14:54:24.000Z","title":"GHSA-8hmm-4crw-vm2c / CVE-2025-57755: @musistudio/claude-code-router has improper CORS configuration","url":"https://github.com/advisories/GHSA-8hmm-4crw-vm2c","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"55b90c70960f","published":"2025-08-21T14:53:52.000Z","title":"GHSA-pp7p-q8fx-2968 / CVE-2025-57753: vite-plugin-static-copy files not included in `src` are possible to access with a crafted request","url":"https://github.com/advisories/GHSA-pp7p-q8fx-2968","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d5e1840c2ad3","published":"2025-08-21T14:48:00.000Z","title":"Microsoft asks customers for feedback on reported SSD failures","url":"https://bleepingcomputer.com/news/microsoft/microsoft-asks-customers-for-feedback-on-ssd-failure-issues","summary":"Microsoft is seeking further information from customers who reported failure and data corruption issues affecting their solid-state drives (SSDs) and hard disk drives (HDDs) after installing the August 2025 security update.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"analysis"},{"id":"c91e4e657a84","published":"2025-08-21T14:47:55.000Z","title":"GHSA-95m3-7q98-8xr5 / CVE-2025-9288: sha.js is missing type checks leading to hash rewind and passing on crafted data","url":"https://github.com/advisories/GHSA-95m3-7q98-8xr5","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"3db46f90e0e0","published":"2025-08-21T14:47:35.000Z","title":"GHSA-cpq7-6gpm-g9rc / CVE-2025-9287: cipher-base is missing type checks, leading to hash rewind and passing on crafted data","url":"https://github.com/advisories/GHSA-cpq7-6gpm-g9rc","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"e89d1750884d","published":"2025-08-21T14:46:51.000Z","title":"GHSA-79j6-g2m3-jgfw / CVE-2025-9141: vLLM has remote code execution vulnerability in the tool call parser for Qwen3-Coder","url":"https://github.com/advisories/GHSA-79j6-g2m3-jgfw","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6b01f276852d","published":"2025-08-21T14:37:19.000Z","title":"GHSA-2464-8j7c-4cjm / CVE-2025-11065: go-viper's mapstructure May Leak Sensitive Information in Logs When Processing Malformed Data","url":"https://github.com/advisories/GHSA-2464-8j7c-4cjm","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"987f7b981d15","published":"2025-08-21T14:27:18.000Z","title":"GHSA-287x-6r2h-f9mw / CVE-2025-55744: UnoPim vulnerable to CSRF on Product edit feature and creation of other types","url":"https://github.com/advisories/GHSA-287x-6r2h-f9mw","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"7dbb3ecc42ce","published":"2025-08-21T14:26:31.000Z","title":"GHSA-v22v-xwh7-2vrm / CVE-2025-55743: UnoPim vulnerable to remote code execution through Arbitrary File upload","url":"https://github.com/advisories/GHSA-v22v-xwh7-2vrm","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"8ee02a9d4553","published":"2025-08-21T14:25:23.000Z","title":"GHSA-xr97-25v7-hc2q / CVE-2025-55742: UnoPim has Stored Cross-site Scripting vulnerability in user creation functionality","url":"https://github.com/advisories/GHSA-xr97-25v7-hc2q","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"f7f741262e97","published":"2025-08-21T14:24:16.000Z","title":"GHSA-rxc4-3w6r-4v47 / CVE-2025-48956: vllm API endpoints vulnerable to Denial of Service Attacks","url":"https://github.com/advisories/GHSA-rxc4-3w6r-4v47","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ecd7dd821af0","published":"2025-08-21T13:53:00.000Z","title":"Why Certified VMware Pros Are Driving the Future of IT","url":"https://bleepingcomputer.com/news/security/why-certified-vmware-pros-are-driving-the-future-of-it","summary":"From hybrid cloud to AI, IT complexity and security risks are accelerating. Certified VMware pros bring clarity and resilience that keep teams ahead. See why CIOs are making certification a workforce strategy with VMUG.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"6afdd41997c9","published":"2025-08-21T13:17:01.000Z","title":"BadFU: Backdoor Federated Learning through Adversarial Machine Unlearning","url":"https://arxiv.org/abs/2508.15541","summary":"Federated learning (FL) has been widely adopted as a decentralized training paradigm that enables multiple clients to collaboratively learn a shared model without exposing their local data. As concerns over data privacy and regulatory compliance grow, machine unlearning, which…","feed":"arxiv-models","publisher":"arxiv","topics":["trust","models"],"type":"research"},{"id":"a542270c97b4","published":"2025-08-21T10:41:00.003Z","title":"Hackers Using New QuirkyLoader Malware to Spread Agent Tesla, AsyncRAT and Snake Keylogger","url":"https://thehackernews.com/2025/08/hackers-using-new-quirkyloader-malware.html","summary":"Cybersecurity researchers have disclosed details of a new malware loader called QuirkyLoader that's being used to deliver via email spam campaigns an array of next-stage payloads ranging from information stealers to remote access trojans since November 2024. Some of the notable…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"research"},{"id":"492c6ab850a7","published":"2025-08-21T10:00:32.000Z","title":"Your Connection, Their Cash: Threat Actors Misuse SDKs to Sell Your Bandwidth","url":"https://unit42.paloaltonetworks.com/attackers-sell-your-bandwidth-using-sdks","summary":"A campaign leverages CVE-2024-36401 to stealthily monetize victims' bandwidth where legitimate software development kits (SDKs) are deployed for passive income.","feed":"unit42","publisher":"palo-alto","topics":[],"type":"analysis"},{"id":"fdb249ed87dd","published":"2025-08-21T09:30:22.000Z","title":"GHSA-x67c-v8jr-p29r / CVE-2025-8023: Mattermost Fails to Sanitize Path Traversal Sequences","url":"https://github.com/advisories/GHSA-x67c-v8jr-p29r","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"08097fbf741c","published":"2025-08-21T09:30:21.000Z","title":"GHSA-4276-cm8c-788h / CVE-2025-53971: Mattermost Fails to Properly Validate Team Role Modification","url":"https://github.com/advisories/GHSA-4276-cm8c-788h","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"3c0e020fd54b","published":"2025-08-21T09:30:21.000Z","title":"GHSA-q453-638c-h4mr / CVE-2025-49222: Mattermost Fails to Validate Remote Cluster Upload Sessions","url":"https://github.com/advisories/GHSA-q453-638c-h4mr","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"642c4fc47a8f","published":"2025-08-21T09:30:21.000Z","title":"GHSA-pwvr-grqg-7vp2 / CVE-2025-49810: Mattermost Lack of Access Control Validation","url":"https://github.com/advisories/GHSA-pwvr-grqg-7vp2","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"788c27468ad0","published":"2025-08-21T09:30:21.000Z","title":"GHSA-vqwh-5jhh-vc9p / CVE-2025-47700: Mattermost Server SSRF Vulnerability via the Agents Plugin","url":"https://github.com/advisories/GHSA-vqwh-5jhh-vc9p","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"8202f63da327","published":"2025-08-21T09:30:21.000Z","title":"GHSA-qj47-w9f2-qg44 / CVE-2025-47870: Mattermost Does Not Sanitize the Team Invite ID","url":"https://github.com/advisories/GHSA-qj47-w9f2-qg44","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a4b5ae0376a5","published":"2025-08-21T09:30:21.000Z","title":"GHSA-gq3r-5833-5532 / CVE-2025-36530: Mattermost Fails to Validate File Paths","url":"https://github.com/advisories/GHSA-gq3r-5833-5532","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"97908028ec35","published":"2025-08-21T09:20:58.000Z","title":"Hijacking Windsurf: How Prompt Injection Leaks Developer Secrets","url":"https://embracethered.com/blog/posts/2025/windsurf-data-exfiltration-vulnerabilities","summary":"This is the first post in a series exploring security vulnerabilities in Windsurf. If you are unfamiliar with Windsurf, it is a fork of VS Code and the coding agent is called Windsurf Cascade . The attack vectors we will explore today allow an adversary during an indirect…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","ai-security"],"type":"incident"},{"id":"2dc7c58f4b54","published":"2025-08-21T07:08:16.000Z","title":"IPIGuard: A Novel Tool Dependency Graph-Based Defense Against Indirect Prompt Injection in LLM Agents","url":"https://arxiv.org/abs/2508.15310","summary":"Large language model (LLM) agents are widely deployed in real-world applications, where they leverage tools to retrieve and manipulate external data for complex tasks. However, when interacting with untrusted data sources (e.g., fetching information from public websites), tool…","feed":"arxiv-sca","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"7997812537c8","published":"2025-08-21T05:25:22.000Z","title":"Retrieval-Augmented Review Generation for Poisoning Recommender Systems","url":"https://arxiv.org/abs/2508.15252","summary":"Recent studies have shown that recommender systems (RSs) are highly vulnerable to data poisoning attacks, where malicious actors inject fake user profiles, including a group of well-designed fake ratings, to manipulate recommendations. Due to security and privacy constraints in…","feed":"arxiv-agentic","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"6617183892a6","published":"2025-08-21T05:21:37.000Z","title":"EMNLP: Educator-role Moral and Normative Large Language Models Profiling","url":"https://arxiv.org/abs/2508.15250","summary":"Simulating Professions (SP) enables Large Language Models (LLMs) to emulate professional roles. However, comprehensive psychological and ethical evaluation in these contexts remains lacking. This paper introduces EMNLP, an Educator-role Moral and Normative LLMs Profiling…","feed":"arxiv-aisec","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"5e280bf3968f","published":"2025-08-21T04:47:00.004Z","title":"Apple Patches CVE-2025-43300 Zero-Day in iOS, iPadOS, and macOS Exploited in Targeted Attacks","url":"https://thehackernews.com/2025/08/apple-patches-cve-2025-43300-zero-day.html","summary":"Apple has released security updates to address a security flaw impacting iOS, iPadOS, and macOS that it said has come under active exploitation in the wild. The zero-day out-of-bounds write vulnerability, tracked as CVE-2025-43300 (CVSS score: 8.8), resides in the ImageIO…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"advisory"},{"id":"80e802c9c907","published":"2025-08-21T02:39:14.000Z","title":"SafeLLM: Unlearning Harmful Outputs from Large Language Models against Jailbreak Attacks","url":"https://arxiv.org/abs/2508.15182","summary":"Jailbreak attacks pose a serious threat to the safety of Large Language Models (LLMs) by crafting adversarial prompts that bypass alignment mechanisms, causing the models to produce harmful, restricted, or biased content. In this paper, we propose SafeLLM, a novel…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"3045c28825fd","published":"2025-08-21T00:30:20.000Z","title":"GHSA-gjx6-h8hm-c9rq / CVE-2025-9264: xxl-job Jobs Handler remove function allows improper control of resource identifiers via ID parameter","url":"https://github.com/advisories/GHSA-gjx6-h8hm-c9rq","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"5d455675c02f","published":"2025-08-21T00:30:20.000Z","title":"GHSA-6rq7-m52p-8pqg / CVE-2025-9263: xxl-job Vulnerable to Resource Injection and Authorization Bypass Through User-Controlled Key","url":"https://github.com/advisories/GHSA-6rq7-m52p-8pqg","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c28572495ff6","published":"2025-08-21T00:30:20.000Z","title":"GHSA-p6rm-483j-37jf / CVE-2025-9262: wong2 mcp-cli Command Injection Vulnerability","url":"https://github.com/advisories/GHSA-p6rm-483j-37jf","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca","ai-security"],"type":"advisory"},{"id":"792af7b9d872","published":"2025-08-21T00:00:00.000Z","title":"MCP vulnerability case study: SQL injection in the Postgres MCP server","url":"https://securitylabs.datadoghq.com/articles/mcp-vulnerability-case-study-SQL-injection-in-the-postgresql-mcp-server","summary":"Learn how vulnerability in Anthropic's reference Postgres MCP server allowed us to bypass the read-only restriction and execute arbitrary SQL statements.","feed":"datadog-seclabs","publisher":"datadog","topics":["ai-security"],"type":"research"},{"id":"95567fa55fc9","published":"2025-08-21T00:00:00.000Z","title":"CVE-2025-43300: Apple iOS, iPadOS, and macOS Out-of-Bounds Write Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2025-43300","summary":"Apple iOS, iPadOS, and macOS contain an out-of-bounds write vulnerability in the Image I/O framework. US federal agencies must act by 2025-09-11.","feed":"cisa-kev","publisher":"cisa","topics":["agentic-sdlc"],"type":"advisory"},{"id":"8d4e92894b54","published":"2025-08-20T23:00:15.000Z","title":"Logit-Gap Steering: A New Frontier in Understanding and Probing LLM Safety","url":"https://unit42.paloaltonetworks.com/logit-gap-steering-impact","summary":"New research from Unit 42 on logit-gap steering reveals how internal alignment measures can be bypassed, making external AI security vital.","feed":"unit42","publisher":"palo-alto","topics":[],"type":"research"},{"id":"ffbe6079eb29","published":"2025-08-20T22:11:00.000Z","title":"AI website builder Lovable increasingly abused for malicious activity","url":"https://bleepingcomputer.com/news/security/ai-website-builder-lovable-increasingly-abused-for-malicious-activity","summary":"Cybercriminals are increasingly abusing the AI-powered Lovable website creation and hosting platform to generate phishing pages, malware-dropping portals, and various fraudulent websites.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["ai-security"],"type":"analysis"},{"id":"c1cf92c74947","published":"2025-08-20T21:30:27.000Z","title":"GHSA-p72g-pv48-7w9x / CVE-2025-54988: Apache Tika XXE Vulnerability via Crafted XFA File Inside a PDF","url":"https://github.com/advisories/GHSA-p72g-pv48-7w9x","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"bce752b4b039","published":"2025-08-20T21:30:26.000Z","title":"GHSA-mpww-r37c-vxjw / CVE-2025-43746: Liferay Portal Vulnerable to Cross-Site Scripting in Dynamic Data Mapping","url":"https://github.com/advisories/GHSA-mpww-r37c-vxjw","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"defe168540a7","published":"2025-08-20T21:30:26.000Z","title":"GHSA-62pf-hcwj-rcfc / CVE-2025-43757: Liferay Portal Vulnerable to Cross-Site Scripting via DDMPortlet_definition Parameter","url":"https://github.com/advisories/GHSA-62pf-hcwj-rcfc","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"4d151d30f985","published":"2025-08-20T20:52:17.000Z","title":"GHSA-mmxm-8w33-wc4h / CVE-2025-5115: Eclipse Jetty affected by MadeYouReset HTTP/2 vulnerability","url":"https://github.com/advisories/GHSA-mmxm-8w33-wc4h","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"81553e78941a","published":"2025-08-20T20:51:55.000Z","title":"GHSA-3j63-5h8p-gf7c: x402 SDK vulnerable in outdated versions in resource servers for builders","url":"https://github.com/advisories/GHSA-3j63-5h8p-gf7c","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"45baf8d3e2f1","published":"2025-08-20T20:09:49.000Z","title":"When Machine Learning Meets Vulnerability Discovery: Challenges and Lessons Learned","url":"https://arxiv.org/abs/2508.15042","summary":"In recent years, machine learning has demonstrated impressive results in various fields, including software vulnerability detection. Nonetheless, using machine learning to identify software vulnerabilities presents new challenges, especially regarding the scale of data…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"f5fb02dedadb","published":"2025-08-20T20:02:35.000Z","title":"MoEcho: Exploiting Side-Channel Attacks to Compromise User Privacy in Mixture-of-Experts LLMs","url":"https://arxiv.org/abs/2508.15036","summary":"The transformer architecture has become a cornerstone of modern AI, fueling remarkable progress across applications in natural language processing, computer vision, and multimodal learning. As these models continue to scale explosively for performance, implementation efficiency…","feed":"arxiv-sca","publisher":"arxiv","topics":[],"type":"research"},{"id":"b9e85e9d0ef5","published":"2025-08-20T19:49:59.000Z","title":"A Systematic Survey of Model Extraction Attacks and Defenses: State-of-the-Art and Perspectives","url":"https://arxiv.org/abs/2508.15031","summary":"Machine learning (ML) models have significantly grown in complexity and utility, driving advances across multiple domains. However, substantial computational resources and specialized expertise have historically restricted their wide adoption. Machine-Learning-as-a-Service…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"8ce4631abad6","published":"2025-08-20T19:33:31.000Z","title":"What is an AI Bill of Materials (AI-BOM)?","url":"https://mend.io/blog/what-is-an-ai-bill-of-materials-ai-bom","summary":"Learn how to create and automate an AI-BOM.","feed":"mend","publisher":"mend","topics":["ai-bom"],"type":"guide"},{"id":"6fc36ffc05c2","published":"2025-08-20T19:09:55.000Z","title":"GHSA-ggjm-f3g4-rwmm / CVE-2025-57749: n8n symlink traversal vulnerability in \"Read/Write File\" node allows access to restricted files","url":"https://github.com/advisories/GHSA-ggjm-f3g4-rwmm","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c8da07a40e69","published":"2025-08-20T19:08:20.000Z","title":"GHSA-mv33-9f6j-pfmc / CVE-2025-55746: Directus allows unauthenticated file upload and file modification due to lacking input sanitization","url":"https://github.com/advisories/GHSA-mv33-9f6j-pfmc","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"5a7e83c9f2c6","published":"2025-08-20T18:44:00.000Z","title":"Apple fixes new zero-day flaw exploited in targeted attacks","url":"https://bleepingcomputer.com/news/apple/apple-emergency-updates-fix-new-actively-exploited-zero-day","summary":"Apple has released emergency updates to patch another zero-day vulnerability that was exploited in an \"extremely sophisticated attack.\"","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"incident"},{"id":"726eafac37cf","published":"2025-08-20T18:30:21.000Z","title":"GHSA-x485-rhg3-cqr4 / CVE-2011-10026: Spree Commerce is vulnerable to RCE through Search API","url":"https://github.com/advisories/GHSA-x485-rhg3-cqr4","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c137533a5834","published":"2025-08-20T17:54:00.002Z","title":"DOM-Based Extension Clickjacking Exposes Popular Password Managers to Credential and Data Theft","url":"https://thehackernews.com/2025/08/dom-based-extension-clickjacking.html","summary":"Popular password manager plugins for web browsers have been found susceptible to clickjacking security vulnerabilities that could be exploited to steal account credentials, two-factor authentication (2FA) codes, and credit card details under certain conditions. The technique…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"59d2f81a046e","published":"2025-08-20T17:25:00.001Z","title":"🕵️ Webinar: Discover and Control Shadow AI Agents in Your Enterprise Before Hackers Do","url":"https://thehackernews.com/2025/08/webinar-discover-and-control-shadow-ai.html","summary":"Do you know how many AI agents are running inside your business right now? If the answer is “not sure,” you’re not alone—and that’s exactly the concern. Across industries, AI agents are being set up every day. Sometimes by IT, but often by business units moving fast to get…","feed":"thehackernews","publisher":"thehackernews","topics":["ai-security"],"type":"analysis"},{"id":"20a3100c5c6e","published":"2025-08-20T17:03:32.000Z","title":"Universal and Transferable Adversarial Attack on Large Language Models Using Exponentiated Gradient Descent","url":"https://arxiv.org/abs/2508.14853","summary":"As large language models (LLMs) are increasingly deployed in critical applications, ensuring their robustness and safety alignment remains a major challenge. Despite the overall success of alignment techniques such as reinforcement learning from human feedback (RLHF) on typical…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"7a6650f45302","published":"2025-08-20T17:00:00.000Z","title":"Mixi reimagines communication with ChatGPT","url":"https://openai.com/index/mixi","summary":"Discover how MIXI, a leader in digital entertainment and lifestyle services in Japan, uses ChatGPT Enterprise to transform productivity, boost AI adoption across teams, and create a secure environment for innovation.","feed":"openai","publisher":"openai","topics":[],"type":"analysis"},{"id":"85288526b61b","published":"2025-08-20T16:31:00.000Z","title":"Perplexity’s Comet AI browser tricked into buying fake items online","url":"https://bleepingcomputer.com/news/security/perplexitys-comet-ai-browser-tricked-into-buying-fake-items-online","summary":"A study looking into agentic AI browsers has found that these emerging tools are vulnerable to both new and old schemes that could make them interact with malicious pages and prompts.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"research"},{"id":"e01b9ba82d2c","published":"2025-08-20T15:59:00.007Z","title":"FBI Warns FSB-Linked Hackers Exploiting Unpatched Cisco Devices for Cyber Espionage","url":"https://thehackernews.com/2025/08/fbi-warns-russian-fsb-linked-hackers.html","summary":"A Russian state-sponsored cyber espionage group known as Static Tundra has been observed actively exploiting a seven-year-old security flaw in Cisco IOS and Cisco IOS XE software as a means to establish persistent access to target networks. Cisco Talos, which disclosed details…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"3a1045c1d81f","published":"2025-08-20T15:31:42.000Z","title":"GHSA-f9qj-4c5x-cpcw / CVE-2025-50864: elysia-cors Origin Validation Error","url":"https://github.com/advisories/GHSA-f9qj-4c5x-cpcw","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"faca1804a04f","published":"2025-08-20T15:31:42.000Z","title":"GHSA-p9gc-59hf-x48p / CVE-2025-43748: Liferay Portal Vulnerable to Cross-Site Request Forgery","url":"https://github.com/advisories/GHSA-p9gc-59hf-x48p","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"41882ca08314","published":"2025-08-20T15:31:41.000Z","title":"GHSA-5fx5-cff6-f3fp / CVE-2025-43749: Liferay Portal Unauthenticated File Access via URL","url":"https://github.com/advisories/GHSA-5fx5-cff6-f3fp","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"9e5afbc7ecba","published":"2025-08-20T15:31:41.000Z","title":"GHSA-56qj-wp5r-mvhj / CVE-2025-43750: Liferay Portal Unvalidated File Upload","url":"https://github.com/advisories/GHSA-56qj-wp5r-mvhj","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a8f476d9ce9d","published":"2025-08-20T15:31:41.000Z","title":"GHSA-8f93-j3fx-72f3 / CVE-2025-4437: CRI-O has Potential High Memory Consumption from File Read","url":"https://github.com/advisories/GHSA-8f93-j3fx-72f3","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"7d934df2fbfb","published":"2025-08-20T14:16:21.000Z","title":"Assessing the Quality and Security of AI-Generated Code: A Quantitative Analysis","url":"https://arxiv.org/abs/2508.14727","summary":"This study presents a quantitative evaluation of the code quality and security of five prominent Large Language Models (LLMs): Claude Sonnet 4, Claude 3.7 Sonnet, GPT-4o, Llama 3.2 90B, and OpenCoder 8B. While prior research has assessed the functional performance of…","feed":"arxiv-agentic","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"cbe9f577c419","published":"2025-08-20T14:01:00.000Z","title":"Why email security needs its EDR moment to move beyond prevention","url":"https://bleepingcomputer.com/news/security/why-email-security-needs-its-edr-moment-to-move-beyond-prevention","summary":"Email security is stuck where antivirus was a decade ago—focused only on prevention. Learn from Material Security why it's time for an \"EDR for email\" mindset: visibility, post-compromise controls, and SaaS-wide protection.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"ec5f39e29869","published":"2025-08-20T13:01:00.007Z","title":"Experts Find AI Browsers Can Be Tricked by PromptFix Exploit to Run Malicious Hidden Prompts","url":"https://thehackernews.com/2025/08/experts-find-ai-browsers-can-be-tricked.html","summary":"Cybersecurity researchers have demonstrated a new prompt injection technique called PromptFix that tricks a generative artificial intelligence (GenAI) model into carrying out unintended actions by embedding the malicious instruction inside a fake CAPTCHA check on a web page.…","feed":"thehackernews","publisher":"thehackernews","topics":["ai-security"],"type":"research"},{"id":"19e862b60b07","published":"2025-08-20T12:31:15.000Z","title":"GHSA-3fp2-6mwq-4q3j / CVE-2025-43742: Liferay Portal Vulnerable to Cross-Site Scripting through URLs","url":"https://github.com/advisories/GHSA-3fp2-6mwq-4q3j","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"242201db653d","published":"2025-08-20T12:31:15.000Z","title":"GHSA-j6p8-g3rj-ghpm / CVE-2025-43741: Liferay Portal Vulnerable to Cross-Site Scripting via assetTagNames Parameter","url":"https://github.com/advisories/GHSA-j6p8-g3rj-ghpm","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"27f63360fc45","published":"2025-08-20T11:02:27.000Z","title":"Subverting AIOps Systems Through Poisoned Input Data","url":"https://schneier.com/blog/archives/2025/08/subverting-aiops-systems-through-poisoned-input-data.html","summary":"In this input integrity attack against an AI system, researchers were able to fool AIOps tools: AIOps refers to the use of LLM-based agents to gather and analyze application telemetry, including system logs, performance metrics, traces, and alerts, to detect problems and then…","feed":"schneier","publisher":"schneier","topics":[],"type":"research"},{"id":"6d72532e515c","published":"2025-08-20T11:00:00.000Z","title":"Amazon Q Developer for VS Code Vulnerable to Invisible Prompt Injection","url":"https://embracethered.com/blog/posts/2025/amazon-q-developer-interprets-hidden-instructions","summary":"The Amazon Q Developer VS Code Extension (Amazon Q) is a very popular coding agent, with over 1 million downloads . In previous posts we showed how prompt injection vulnerabilities in Amazon Q could lead to: Exfiltration of sensitive information from the user’s machine , and…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","ai-security"],"type":"analysis"},{"id":"52b147e0a6c6","published":"2025-08-20T09:30:41.000Z","title":"GHSA-hf86-8x8v-h7vc / CVE-2024-39954: Apache EventMesh Vulnerable to Server-Side Request Forgery in WebhookUtil.java","url":"https://github.com/advisories/GHSA-hf86-8x8v-h7vc","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"dc2b4970e6cb","published":"2025-08-20T08:39:12.000Z","title":"DOPA: Stealthy and Generalizable Backdoor Attacks from a Single Client under Challenging Federated Constraints","url":"https://arxiv.org/abs/2508.14530","summary":"Federated Learning (FL) is increasingly adopted for privacy-preserving collaborative training, but its decentralized nature makes it particularly susceptible to backdoor attacks. Existing attack methods, however, often rely on idealized assumptions and fail to remain effective…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"63f3e0c21d16","published":"2025-08-20T06:46:00.000Z","title":"Microsoft releases emergency updates to fix Windows recovery","url":"https://bleepingcomputer.com/news/microsoft/microsoft-releases-emergency-updates-to-fix-windows-recovery","summary":"Microsoft has released emergency Windows out-of-band updates to resolve a known issue breaking reset and recovery operations after installing the August 2025 Windows security updates.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"research"},{"id":"4d3c9dbad35b","published":"2025-08-19T22:57:17.000Z","title":"Mechanistic Exploration of Backdoored Large Language Model Attention Patterns","url":"https://arxiv.org/abs/2508.15847","summary":"Backdoor attacks creating 'sleeper agents' in large language models (LLMs) pose significant safety risks. This study employs mechanistic interpretability to explore resulting internal structural differences. Comparing clean Qwen2.5-3B models with versions poisoned using…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"7b4cdcd5932e","published":"2025-08-19T22:24:40.000Z","title":"GHSA-pr72-8fxw-xx22 / CVE-2025-55740: Default Credentials in nginx-defender Configuration Files","url":"https://github.com/advisories/GHSA-pr72-8fxw-xx22","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"48eff8952d5d","published":"2025-08-19T21:30:37.000Z","title":"GHSA-m49p-6cjp-x2h3 / CVE-2025-43744: Liferay Portal Vulnerable to Cross-Site Scripting via DDM Structure Field Labels","url":"https://github.com/advisories/GHSA-m49p-6cjp-x2h3","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c7198412fa24","published":"2025-08-19T21:30:37.000Z","title":"GHSA-g4vp-4gqr-7v8c / CVE-2025-43743: Liferay Portal Enumeration Discrepancy in Calendars","url":"https://github.com/advisories/GHSA-g4vp-4gqr-7v8c","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"2588f3462a4b","published":"2025-08-19T21:30:36.000Z","title":"GHSA-7q33-gwcm-r6cj / CVE-2025-43745: Liferay Portal CSRF Vulnerability via Endpoint Parameter","url":"https://github.com/advisories/GHSA-7q33-gwcm-r6cj","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c1d292ab6634","published":"2025-08-19T21:30:36.000Z","title":"GHSA-vjwr-cqwf-6q96 / CVE-2025-43737: Liferay Portal Vulnerable to Cross-Site Scripting via backURL Paramter","url":"https://github.com/advisories/GHSA-vjwr-cqwf-6q96","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c728220a426f","published":"2025-08-19T21:20:58.000Z","title":"Amazon Q Developer: Remote Code Execution with Prompt Injection","url":"https://embracethered.com/blog/posts/2025/amazon-q-developer-remote-code-execution","summary":"The Amazon Q Developer VS Code Extension (Amazon Q) is a popular coding agent, with over 1 million downloads . The extension is vulnerable to indirect prompt injection, and in this post we discuss a vulnerability that allowed an adversary (or also the AI for that matter) to run…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","remediation","ai-security"],"type":"analysis"},{"id":"df6e22be6109","published":"2025-08-19T20:41:10.000Z","title":"GHSA-35c5-67fm-cpcp / CVE-2025-8678: WP Crontrol Authenticated (Administrator+) plugin vulnerable to Blind Server-Side Request Forgery","url":"https://github.com/advisories/GHSA-35c5-67fm-cpcp","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"8f105acbf422","published":"2025-08-19T20:17:45.000Z","title":"GHSA-gjx4-2c7g-fm94 / CVE-2025-55294: screenshot-desktop vulnerable to command Injection via `format` option","url":"https://github.com/advisories/GHSA-gjx4-2c7g-fm94","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"55f800fa9011","published":"2025-08-19T20:16:58.000Z","title":"GHSA-7rqq-prvp-x9jh / CVE-2025-54881: Mermaid improperly sanitizes sequence diagram labels leading to XSS","url":"https://github.com/advisories/GHSA-7rqq-prvp-x9jh","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"391462adb567","published":"2025-08-19T20:16:45.000Z","title":"GHSA-8gwm-58g9-j8pw / CVE-2025-54880: Mermaid does not properly sanitize architecture diagram iconText leading to XSS","url":"https://github.com/advisories/GHSA-8gwm-58g9-j8pw","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"25e96ad61650","published":"2025-08-19T20:08:00.000Z","title":"PyPI now blocks domain resurrection attacks used for hijacking accounts","url":"https://bleepingcomputer.com/news/security/pypi-now-blocks-domain-resurrection-attacks-used-for-hijacking-accounts","summary":"The Python Package Index (PyPI) has introduced new protections against domain resurrection attacks that enable hijacking accounts through password resets.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["sca"],"type":"analysis"},{"id":"253dd473a7e6","published":"2025-08-19T19:39:37.000Z","title":"Incident Analysis for AI Agents","url":"https://arxiv.org/abs/2508.14231","summary":"As AI agents become more widely deployed, we are likely to see an increasing number of incidents: events involving AI agent use that directly or indirectly cause harm. For example, agents could be prompt-injected to exfiltrate private information or make unauthorized purchases.…","feed":"arxiv-aisec","publisher":"arxiv","topics":[],"type":"research"},{"id":"84d38223b95b","published":"2025-08-19T18:31:31.000Z","title":"GHSA-cwgh-r52j-xh6c / CVE-2025-43738: Liferay Portal Reflected Cross-Site Scripting Vulnerability in displayType Parameter","url":"https://github.com/advisories/GHSA-cwgh-r52j-xh6c","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b8b9d9e0f4f1","published":"2025-08-19T18:07:00.000Z","title":"Microsoft shares workaround for Teams \"couldn't connect\" error","url":"https://bleepingcomputer.com/news/microsoft/microsoft-shares-workaround-for-teams-couldnt-connect-error","summary":"Microsoft is resolving a known issue that causes \"couldn't connect\" errors when launching the Microsoft Teams desktop and web applications.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"analysis"},{"id":"3cdb628e83e3","published":"2025-08-19T17:37:00.000Z","title":"Apache ActiveMQ Flaw Exploited to Deploy DripDropper Malware on Cloud Linux Systems","url":"https://thehackernews.com/2025/08/apache-activemq-flaw-exploited-to.html","summary":"Threat actors are exploiting a nearly two-year-old security flaw in Apache ActiveMQ to gain persistent access to cloud Linux systems and deploy malware called DripDropper. But in an unusual twist, the unknown attackers have been observed patching the exploited vulnerability…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"incident"},{"id":"ba9a3e22ed19","published":"2025-08-19T16:41:00.000Z","title":"Elastic rejects claims of a zero-day RCE flaw in Defend EDR","url":"https://bleepingcomputer.com/news/security/elastic-rejects-claims-of-a-zero-day-rce-flaw-in-defend-edr","summary":"Enterprise search and security company Elastic is rejecting reports of a zero-day vulnerability impacting its Defend endpoint detection and response (EDR) product.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"2679472ef827","published":"2025-08-19T16:39:31.000Z","title":"What is Generative AI Security?","url":"https://mend.io/blog/what-is-generative-ai-security","summary":"Learn what generative AI in cybersecurity is and how to secure against threats.","feed":"mend","publisher":"mend","topics":[],"type":"guide"},{"id":"cc920ba8d2ce","published":"2025-08-19T15:40:31.000Z","title":"GHSA-xf8x-j4p2-f749 / CVE-2025-55303: Astro allows unauthorized third-party images in _image endpoint","url":"https://github.com/advisories/GHSA-xf8x-j4p2-f749","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"0a628db87547","published":"2025-08-19T15:36:44.000Z","title":"PyPI: Preventing Domain Resurrection Attacks","url":"https://simonwillison.net/2025/Aug/19/pypi-preventing-domain-resurrection-attacks","summary":"PyPI: Preventing Domain Resurrection Attacks Domain resurrection attacks are a nasty vulnerability in systems that use email verification to allow people to recover their accounts. If somebody lets their domain name expire an attacker might snap it up and use it to gain access…","feed":"simonw","publisher":"simonw","topics":["sca"],"type":"analysis"},{"id":"c6e0c92c89e9","published":"2025-08-19T15:34:48.000Z","title":"GHSA-qp7j-x725-g67f: HydrAIDE Authentication Bypass Vulnerability","url":"https://github.com/advisories/GHSA-qp7j-x725-g67f","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b1f12ba5f5ce","published":"2025-08-19T15:33:35.000Z","title":"GHSA-hfmv-hhh3-43f2 / CVE-2025-52478: Stored XSS in n8n Form Trigger allows Account Takeover via injected iframe and video/source","url":"https://github.com/advisories/GHSA-hfmv-hhh3-43f2","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"028a9b229ef0","published":"2025-08-19T15:31:29.000Z","title":"GHSA-rh9f-gr6q-mpc4 / CVE-2025-51488: moonshine Stored Cross-Site Scripting Vulnerability in Create Admin","url":"https://github.com/advisories/GHSA-rh9f-gr6q-mpc4","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"1991b35a5259","published":"2025-08-19T15:31:29.000Z","title":"GHSA-p632-58pp-c9xg / CVE-2025-51487: moonshine Stored Cross-Site Scripting Vulnerability in Create Article","url":"https://github.com/advisories/GHSA-p632-58pp-c9xg","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"9d8d0b880930","published":"2025-08-19T15:31:29.000Z","title":"GHSA-9g9j-3w64-3cjh / CVE-2025-51510: MoonShine SQL Injection Vulnerability","url":"https://github.com/advisories/GHSA-9g9j-3w64-3cjh","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"fc82313fa2f2","published":"2025-08-19T15:31:29.000Z","title":"GHSA-8xfq-7f6m-mpmf / CVE-2025-51489: MoonShine Arbitrary File Upload Vulnerability","url":"https://github.com/advisories/GHSA-8xfq-7f6m-mpmf","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"179cd786923e","published":"2025-08-19T15:31:28.000Z","title":"GHSA-22jp-w3cg-gvmm / CVE-2025-43740: Liferay Portal has Stored Cross-Site Scripting Vulnerability via Message Boards Feature","url":"https://github.com/advisories/GHSA-22jp-w3cg-gvmm","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"db88829c2554","published":"2025-08-19T15:31:28.000Z","title":"GHSA-7mxq-h2r7-h449 / CVE-2025-43739: Liferay Portal Email Modification Vulnerability via Calendar Portlet","url":"https://github.com/advisories/GHSA-7mxq-h2r7-h449","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"0bc825e8a9b7","published":"2025-08-19T14:14:58.000Z","title":"FedUP: Efficient Pruning-based Federated Unlearning for Model Poisoning Attacks","url":"https://arxiv.org/abs/2508.13853","summary":"Federated Learning (FL) can be vulnerable to attacks, such as model poisoning, where adversaries send malicious local weights to compromise the global model. Federated Unlearning (FU) is emerging as a solution to address such vulnerabilities by selectively removing the…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"39c867c854ad","published":"2025-08-19T13:56:09.000Z","title":"CIA+TA Risk Assessment for AI Reasoning Vulnerabilities","url":"https://arxiv.org/abs/2508.15839","summary":"As AI systems increasingly influence critical decisions, they face threats that exploit reasoning mechanisms rather than technical infrastructure. We present a framework for cognitive cybersecurity, a systematic protection of AI reasoning processes from adversarial…","feed":"arxiv-sca","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"026d445423e1","published":"2025-08-19T13:24:46.000Z","title":"Structural and Connectivity Patterns in the Maven Central Software Dependency Network","url":"https://arxiv.org/abs/2508.13819","summary":"Understanding the structural characteristics and connectivity patterns of large-scale software ecosystems is critical for enhancing software reuse, improving ecosystem resilience, and mitigating security risks. In this paper, we investigate the Maven Central ecosystem, one of…","feed":"arxiv-sca","publisher":"arxiv","topics":[],"type":"research"},{"id":"4c21cb2ffa5e","published":"2025-08-19T13:00:00.001Z","title":"Public Exploit for Chained SAP Flaws Exposes Unpatched Systems to Remote Code Execution","url":"https://thehackernews.com/2025/08/public-exploit-for-chained-sap-flaws.html","summary":"A new exploit combining two critical, now-patched security flaws in SAP NetWeaver has emerged in the wild, putting organizations at risk of system compromise and data theft. The exploit in question chains together CVE-2025-31324 and CVE-2025-42999 to bypass authentication and…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"analysis"},{"id":"5e21eab65b2d","published":"2025-08-19T12:00:01.000Z","title":"Wiz Completes IRAP Assessment to Support Australian Government Cloud Security","url":"https://wiz.io/blog/irap-assessment-australia-public-sector","summary":"Empowering Australian government agencies with enhanced cloud security","feed":"wiz","publisher":"wiz","topics":[],"type":"analysis"},{"id":"ad04ea2e9783","published":"2025-08-19T11:43:06.000Z","title":"NodeShield: Runtime Enforcement of Security-Enhanced SBOMs for Node.js","url":"https://arxiv.org/abs/2508.13750","summary":"The software supply chain is an increasingly common attack vector for malicious actors. The Node.js ecosystem has been subject to a wide array of attacks, likely due to its size and prevalence. To counter such attacks, the research community and practitioners have proposed a…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast","sca"],"type":"research"},{"id":"e2cc1ced26cb","published":"2025-08-19T10:12:35.000Z","title":"MCPTox: A Benchmark for Tool Poisoning Attack on Real-World MCP Servers","url":"https://arxiv.org/abs/2508.14925","summary":"By providing a standardized interface for LLM agents to interact with external tools, the Model Context Protocol (MCP) is quickly becoming a cornerstone of the modern autonomous agent ecosystem. However, it creates novel attack surfaces due to untrusted external tools. While…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"dee89b0b99df","published":"2025-08-19T06:36:00.002Z","title":"PyPI Blocks 1,800 Expired-Domain Emails to Prevent Account Takeovers and Supply Chain Attacks","url":"https://thehackernews.com/2025/08/pypi-blocks-1800-expired-domain-emails.html","summary":"The maintainers of the Python Package Index (PyPI) repository have announced that the package manager now checks for expired domains to prevent supply chain attacks. \"These changes improve PyPI's overall account security posture, making it harder for attackers to exploit…","feed":"thehackernews","publisher":"thehackernews","topics":["sca"],"type":"analysis"},{"id":"6f338a16bc59","published":"2025-08-19T06:00:00.000Z","title":"Last Week in AppSec for 19. August 2025","url":"https://checkmarx.com/zero-post/last-week-in-appsec-2025-08-19","summary":"code injection in AI Agent dev tool, path traversal in `go-getter`, model code injection protection bypass in TensorFlow Keras, and unsafe ImageMagick use in Rails Active storage — last week in AppSec","feed":"checkmarx","publisher":"checkmarx","topics":["ai-security"],"type":"analysis"},{"id":"7d416f625599","published":"2025-08-19T05:05:12.000Z","title":"GHSA-64xw-qfrf-5238: Malware in monitor-package","url":"https://github.com/advisories/GHSA-64xw-qfrf-5238","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-malware","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a07eef542b6c","published":"2025-08-19T05:02:35.000Z","title":"GHSA-23p5-xx6f-c4wh: Malware in grok-sdk","url":"https://github.com/advisories/GHSA-23p5-xx6f-c4wh","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-malware","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ddfdfd20437d","published":"2025-08-19T04:17:21.000Z","title":"CCFC: Core & Core-Full-Core Dual-Track Defense for LLM Jailbreak Protection","url":"https://arxiv.org/abs/2508.14128","summary":"Jailbreak attacks pose a serious challenge to the safe deployment of large language models (LLMs). We introduce CCFC (Core & Core-Full-Core), a dual-track, prompt-level defense framework designed to mitigate LLMs' vulnerabilities from prompt injection and structure-aware…","feed":"arxiv-models","publisher":"arxiv","topics":["models","ai-security"],"type":"research"},{"id":"95a890ea52eb","published":"2025-08-19T02:17:48.000Z","title":"Beneath the Mask: Can Contribution Data Unveil Malicious Personas in Open-Source Projects?","url":"https://arxiv.org/abs/2508.13453","summary":"In February 2024, after building trust over two years with project maintainers by making a significant volume of legitimate contributions, GitHub user \"JiaT75\" self-merged a version of the XZ Utils project containing a highly sophisticated, well-disguised backdoor targeting…","feed":"arxiv-models","publisher":"arxiv","topics":["sca"],"type":"research"},{"id":"8c7a194efbcc","published":"2025-08-18T22:25:29.000Z","title":"GHSA-vxq6-8cwm-wj99 / CVE-2025-55296: LibreNMS allows stored XSS in Alert Template name field","url":"https://github.com/advisories/GHSA-vxq6-8cwm-wj99","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c4e58f3902e4","published":"2025-08-18T22:00:00.000Z","title":"Trivial C# Random Exploitation","url":"https://blog.doyensec.com/2025/08/19/trivial-exploit-on-C-random.html","summary":"Exploiting random number generators requires math, right? Thanks to C#’s Random , that is not necessarily the case! I ran into an HTTP 2.0 web service issuing password reset tokens from a custom encoding of (new Random()).Next(min, max) output. This led to a critical account…","feed":"doyensec","publisher":"doyensec","topics":["dast"],"type":"analysis"},{"id":"aab0671d6322","published":"2025-08-18T21:31:18.000Z","title":"GHSA-3p2m-574v-v257 / CVE-2025-43731: Liferay Portal Vulnerable to Cross-Site Scripting","url":"https://github.com/advisories/GHSA-3p2m-574v-v257","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"672637b70941","published":"2025-08-18T21:01:07.000Z","title":"GHSA-p7q8-grrj-3m8w / CVE-2025-55214: Copier's safe template has filesystem write access outside destination path","url":"https://github.com/advisories/GHSA-p7q8-grrj-3m8w","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"56a4f65de4eb","published":"2025-08-18T21:00:52.000Z","title":"GHSA-mgh9-4mwp-fg55 / CVE-2025-55213: OpenFGA Authorization Bypass ","url":"https://github.com/advisories/GHSA-mgh9-4mwp-fg55","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"891f17c2343d","published":"2025-08-18T21:00:36.000Z","title":"GHSA-fcpm-6mxq-m5vv / CVE-2025-55205: Capsule tenant owners with \"patch namespace\" permission can hijack system namespaces label","url":"https://github.com/advisories/GHSA-fcpm-6mxq-m5vv","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca","remediation"],"type":"advisory"},{"id":"042d8445cf31","published":"2025-08-18T21:00:23.000Z","title":"GHSA-3xw7-v6cj-5q8h / CVE-2025-55201: Copier's safe template has arbitrary filesystem read/write access","url":"https://github.com/advisories/GHSA-3xw7-v6cj-5q8h","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"eee847e94976","published":"2025-08-18T19:20:58.000Z","title":"Amazon Q Developer: Secrets Leaked via DNS and Prompt Injection","url":"https://embracethered.com/blog/posts/2025/amazon-q-developer-data-exfil-via-dns","summary":"The next three posts will cover high severity vulnerabilities in the Amazon Q Developer VS Code Extension (Amazon Q Developer), which is a very popular coding agent, with over 1 million downloads. It is vulnerable to prompt injection from untrusted data and its security depends…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["sca","agentic-sdlc","ai-security"],"type":"incident"},{"id":"df8404e2b321","published":"2025-08-18T18:46:52.000Z","title":"GHSA-x5gv-jw7f-j6xj / CVE-2025-55284: Claude Code's Permissive Default Allowlist Enables Unauthorized File Read and Network Exfiltration in Claude Code","url":"https://github.com/advisories/GHSA-x5gv-jw7f-j6xj","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca","agentic-sdlc"],"type":"advisory"},{"id":"792f297e3f0d","published":"2025-08-18T18:30:35.000Z","title":"GHSA-g4wg-mpfg-x2q6 / CVE-2025-3639: Liferay Portal Login Bypass Vulnerability","url":"https://github.com/advisories/GHSA-g4wg-mpfg-x2q6","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"9c3206209ac5","published":"2025-08-18T18:12:00.000Z","title":"ERMAC Android malware source code leak exposes banking trojan infrastructure","url":"https://bleepingcomputer.com/news/security/ermac-android-malware-source-code-leak-exposes-banking-trojan-infrastructure","summary":"The source code for version 3 of the ERMAC Android banking trojan has been leaked online, exposing the internals of the malware-as-a-service platform and the operator's infrastructure.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"46a07dbaf767","published":"2025-08-18T17:05:18.000Z","title":"VerilogLAVD: LLM-Aided Rule Generation for Vulnerability Detection in Verilog","url":"https://arxiv.org/abs/2508.13092","summary":"Timely detection of hardware vulnerabilities during the early design stage is critical for reducing remediation costs. Existing early detection techniques often require specialized security expertise, limiting their usability. Recent efforts have explored the use of large…","feed":"arxiv-crse","publisher":"arxiv","topics":[],"type":"research"},{"id":"e9e2f8f4e780","published":"2025-08-18T16:09:57.000Z","title":"MAJIC: Markovian Adaptive Jailbreaking via Iterative Composition of Diverse Innovative Strategies","url":"https://arxiv.org/abs/2508.13048","summary":"Large Language Models (LLMs) have exhibited remarkable capabilities but remain vulnerable to jailbreaking attacks, which can elicit harmful content from the models by manipulating the input prompts. Existing black-box jailbreaking techniques primarily rely on static prompts…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"fbd2b2b0721d","published":"2025-08-18T16:06:00.000Z","title":"Over 800 N-able servers left unpatched against critical flaws","url":"https://bleepingcomputer.com/news/security/over-800-n-able-servers-left-unpatched-against-critical-flaws","summary":"Over 800 N-able N-central servers remain unpatched against a pair of critical security vulnerabilities tagged as actively exploited last week.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"8aaf6e88f591","published":"2025-08-18T16:03:00.003Z","title":"Microsoft Windows Vulnerability Exploited to Deploy PipeMagic RansomExx Malware","url":"https://thehackernews.com/2025/08/microsoft-windows-vulnerability.html","summary":"Cybersecurity researchers have lifted the lid on the threat actors' exploitation of a now-patched security flaw in Microsoft Windows to deploy the PipeMagic malware in RansomExx ransomware attacks. The attacks involve the exploitation of CVE-2025-29824, a privilege escalation…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"incident"},{"id":"4cf824053815","published":"2025-08-18T15:46:36.000Z","title":"The Application of Transformer-Based Models for Predicting Consequences of Cyber Attacks","url":"https://arxiv.org/abs/2508.13030","summary":"Cyberattacks are increasing, and securing against such threats is costing industries billions of dollars annually. Threat Modeling, that is, comprehending the consequences of these attacks, can provide critical support to cybersecurity professionals, enabling them to take…","feed":"arxiv-sca","publisher":"arxiv","topics":[],"type":"research"},{"id":"0646512c89eb","published":"2025-08-18T15:30:32.000Z","title":"GHSA-vhcr-hgc8-29qr / CVE-2025-43733: Liferay Portal Vulnerable to Cross-Site Scripting ","url":"https://github.com/advisories/GHSA-vhcr-hgc8-29qr","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"37d11d70f3fb","published":"2025-08-18T15:30:32.000Z","title":"GHSA-v6xr-v2qg-h22h / CVE-2025-43732: Liferay Portal Vulnerable to Insecure Direct Object Reference","url":"https://github.com/advisories/GHSA-v6xr-v2qg-h22h","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"65d11acf163b","published":"2025-08-18T15:10:15.000Z","title":"GHSA-qq4c-hm99-979m: IdMap from_iter may lead to uninitialized memory being freed on drop","url":"https://github.com/advisories/GHSA-qq4c-hm99-979m","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"68d354d4600b","published":"2025-08-18T15:00:00.000Z","title":"Dissecting PipeMagic: Inside the architecture of a modular backdoor framework","url":"https://microsoft.com/en-us/security/blog/2025/08/18/dissecting-pipemagic-inside-the-architecture-of-a-modular-backdoor-framework","summary":"A comprehensive technical deep dive on PipeMagic, a highly modular backdoor used by Storm-2460 masquerading as a legitimate open-source ChatGPT Desktop Application. Beneath its disguise, PipeMagic is a sophisticated malware framework designed for flexibility and persistence.…","feed":"microsoft-security","publisher":"microsoft","topics":[],"type":"analysis"},{"id":"3e5ebab2867a","published":"2025-08-18T14:24:27.000Z","title":"Documenting Deployment with Fabric: A Repository of Real-World AI Governance","url":"https://arxiv.org/abs/2508.14119","summary":"Artificial intelligence (AI) is increasingly integrated into society, from financial services and traffic management to creative writing. Academic literature on the deployment of AI has mostly focused on the risks and harms that result from the use of AI. We introduce Fabric, a…","feed":"arxiv-trust","publisher":"arxiv","topics":["trust"],"type":"research"},{"id":"229dfe9ccfb5","published":"2025-08-18T12:54:16.000Z","title":"RAJ-PGA: Reasoning-Activated Jailbreak and Principle-Guided Alignment Framework for Large Reasoning Models","url":"https://arxiv.org/abs/2508.12897","summary":"Large Reasoning Models (LRMs) face a distinct safety vulnerability: their internal reasoning chains may generate harmful content even when the final output appears benign. To address this overlooked risk, we first propose a novel attack paradigm, Reasoning-Activated Jailbreak…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"4d8e1be8e76a","published":"2025-08-18T10:56:00.001Z","title":"Malicious PyPI and npm Packages Discovered Exploiting Dependencies in Supply Chain Attacks","url":"https://thehackernews.com/2025/08/malicious-pypi-and-npm-packages.html","summary":"Cybersecurity researchers have discovered a malicious package in the Python Package Index (PyPI) repository that introduces malicious behavior through a dependency that allows it to establish persistence and achieve code execution. The package, named termncolor, realizes its…","feed":"thehackernews","publisher":"thehackernews","topics":["sca"],"type":"research"},{"id":"a973a8a5120d","published":"2025-08-18T10:38:30.000Z","title":"Involuntary Jailbreak: On Self-Prompting Attacks","url":"https://arxiv.org/abs/2508.13246","summary":"In this study, we disclose a worrying new vulnerability in Large Language Models (LLMs), which we term \\textbf{involuntary jailbreak}. Unlike existing jailbreak attacks, this weakness is distinct in that it does not involve a specific attack objective, such as generating…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"a4edac2c7be4","published":"2025-08-18T09:31:44.000Z","title":"GHSA-r936-gwx5-v52f / CVE-2025-41242: Spring Framework MVC Applications Path Traversal Vulnerability","url":"https://github.com/advisories/GHSA-r936-gwx5-v52f","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a79b5a34ea7d","published":"2025-08-18T02:39:02.000Z","title":"Reducing False Positives with Active Behavioral Analysis for Cloud Security","url":"https://arxiv.org/abs/2508.12584","summary":"Rule-based cloud security posture management (CSPM) solutions are known to produce a lot of false positives based on the limited contextual understanding and dependence on static heuristics testing. This paper introduces a validation-driven methodology that integrates active…","feed":"arxiv-sca","publisher":"arxiv","topics":["dast","models"],"type":"research"},{"id":"b3b42dc85f66","published":"2025-08-18T00:48:50.000Z","title":"XAMT: Cross-Framework API Matching for Testing Deep Learning Libraries","url":"https://arxiv.org/abs/2508.12546","summary":"Deep learning powers critical applications such as autonomous driving, healthcare, and finance, where the correctness of underlying libraries is essential. Bugs in widely used deep learning APIs can propagate to downstream systems, causing serious consequences. While existing…","feed":"arxiv-dast","publisher":"arxiv","topics":["dast"],"type":"research"},{"id":"3fea8930fb83","published":"2025-08-18T00:30:30.000Z","title":"GHSA-q4rg-7cjj-5r86 / CVE-2025-9095: ExpressGateway Cross-Site Scripting Vulnerability in lib/rest/routes/users.js","url":"https://github.com/advisories/GHSA-q4rg-7cjj-5r86","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"9485ef78b64c","published":"2025-08-18T00:30:30.000Z","title":"GHSA-xfp8-x3j6-h67v / CVE-2025-9096: ExpressGateway Cross-Site Scripting Vulnerability in lib/rest/routes/apps.js","url":"https://github.com/advisories/GHSA-xfp8-x3j6-h67v","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6e5ff4306ac7","published":"2025-08-18T00:23:41.000Z","title":"MCPXKIT: The Unified Toolkit for Analyzing Model Context Protocol Security","url":"https://arxiv.org/abs/2508.12538","summary":"The Model Context Protocol (MCP) has emerged as a universal standard that enables AI agents to seamlessly connect with external tools, significantly enhancing their functionality. However, while MCP brings notable benefits, it also introduces significant vulnerabilities, such…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"07a38f2435ee","published":"2025-08-18T00:01:42.000Z","title":"CorrSteer: Generation-Time LLM Steering via Correlated Sparse Autoencoder Features","url":"https://arxiv.org/abs/2508.12535","summary":"Sparse Autoencoders (SAEs) can extract interpretable features from large language models (LLMs) without supervision. However, their effectiveness in downstream steering tasks is limited by the requirement for contrastive datasets or large activation storage. To address these…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"9affa908606e","published":"2025-08-18T00:00:00.000Z","title":"Preventing Domain Resurrection Attacks","url":"https://blog.pypi.org/posts/2025-08-18-preventing-domain-resurrections","summary":"PyPI now checks for expired domains to prevent domain resurrection attacks, a type of supply-chain attack where someone buys an expired domain and uses it to take over PyPI accounts through password resets.","feed":"pypi-blog","publisher":"pypi","topics":["sca"],"type":"analysis"},{"id":"f5f85cc6603f","published":"2025-08-18T00:00:00.000Z","title":"CVE-2025-54948: Trend Micro Apex One OS Command Injection Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2025-54948","summary":"Trend Micro Apex One Management Console (on-premise) contains an OS command injection vulnerability that could allow a pre-authenticated remote attacker to upload malicious code and execute commands on affected installations. US federal agencies must act by 2025-09-08.","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"303963a7161d","published":"2025-08-17T18:50:23.000Z","title":"A Robust Cross-Domain IDS using BiGRU-LSTM-Attention for Medical and Industrial IoT Security","url":"https://arxiv.org/abs/2508.12470","summary":"The increased Internet of Medical Things IoMT and the Industrial Internet of Things IIoT interconnectivity has introduced complex cybersecurity challenges, exposing sensitive data, patient safety, and industrial operations to advanced cyber threats. To mitigate these risks,…","feed":"arxiv-sca","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"49ef383ae7c6","published":"2025-08-17T14:47:31.000Z","title":"ViT-EnsembleAttack: Augmenting Ensemble Models for Stronger Adversarial Transferability in Vision Transformers","url":"https://arxiv.org/abs/2508.12384","summary":"Ensemble-based attacks have been proven to be effective in enhancing adversarial transferability by aggregating the outputs of models with various architectures. However, existing research primarily focuses on refining ensemble weights or optimizing the ensemble path,…","feed":"arxiv-trust","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"264e6c3d3a47","published":"2025-08-17T11:49:16.000Z","title":"MCPSecBench: A Systematic Security Benchmark and Playground for Testing Model Context Protocols","url":"https://arxiv.org/abs/2508.13220","summary":"Large Language Models (LLMs) are increasingly integrated into real-world applications via the Model Context Protocol (MCP), a universal open standard for connecting AI agents with data sources and external tools. While MCP enhances the capabilities of LLM-based agents, it also…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"f1857c857132","published":"2025-08-17T11:10:58.000Z","title":"Data Exfiltration via Image Rendering Fixed in Amp Code","url":"https://embracethered.com/blog/posts/2025/amp-code-fixed-data-exfiltration-via-images","summary":"In this post we discuss a vulnerability that was present in Amp Code from Sourcegraph by which an attacker could exploit markdown driven image rendering to exfiltrate sensitive information. This vulnerability is common in AI applications and agents, and it’s actually similar to…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","remediation","ai-security"],"type":"analysis"},{"id":"dd059f599ab1","published":"2025-08-17T06:52:39.000Z","title":"Fortifying the Agentic Web: A Unified Zero-Trust Architecture Against Logic-layer Threats","url":"https://arxiv.org/abs/2508.12259","summary":"This paper presents a Unified Security Architecture that fortifies the Agentic Web through a Zero-Trust IAM framework. This architecture is built on a foundation of rich, verifiable agent identities using Decentralized Identifiers (DIDs) and Verifiable Credentials (VCs), with…","feed":"arxiv-agentic","publisher":"arxiv","topics":[],"type":"research"},{"id":"5e3e00c74464","published":"2025-08-16T23:30:08.000Z","title":"Too Easily Fooled? Prompt Injection Breaks LLMs on Frustratingly Simple Multiple-Choice Questions","url":"https://arxiv.org/abs/2508.13214","summary":"Large Language Models (LLMs) have recently demonstrated strong emergent abilities in complex reasoning and zero-shot generalization, showing unprecedented potential for LLM-as-a-judge applications in education, peer review, and data quality evaluation. However, their robustness…","feed":"arxiv-aisec","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"443670e348a9","published":"2025-08-16T22:56:51.000Z","title":"Invitation Is All You Need! Promptware Attacks Against LLM-Powered Assistants in Production Are Practical and Dangerous","url":"https://arxiv.org/abs/2508.12175","summary":"The growing integration of LLMs into applications has introduced new security risks, notably known as Promptware - maliciously engineered prompts designed to manipulate LLMs to compromise the CIA triad of these applications. While prior research warned about a potential shift…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"3e1ebcdf4fc1","published":"2025-08-16T19:20:58.000Z","title":"Amp Code: Invisible Prompt Injection Fixed by Sourcegraph","url":"https://embracethered.com/blog/posts/2025/amp-code-fixed-invisible-prompt-injection","summary":"In this post we will look at Amp, a coding agent from Sourcegraph. The other day we discussed how invisible instructions impact Google Jules . Turns out that many client applications are vulnerable to these kinds of attacks when they use models that support invisible…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","ai-security"],"type":"analysis"},{"id":"c6dbd2499459","published":"2025-08-16T15:03:33.000Z","title":"Mitigating Jailbreaks with Intent-Aware LLMs","url":"https://arxiv.org/abs/2508.12072","summary":"Despite extensive safety-tuning, large language models (LLMs) remain vulnerable to jailbreak attacks via adversarially crafted instructions, reflecting a persistent trade-off between safety and task performance. In this work, we propose Intent-FT, a simple and lightweight…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"7ea2e8aee6f0","published":"2025-08-16T14:24:00.000Z","title":"Researcher to release exploit for full auth bypass on FortiWeb","url":"https://bleepingcomputer.com/news/security/researcher-to-release-exploit-for-full-auth-bypass-on-fortiweb","summary":"A security researcher has released a partial proof of concept exploit for a vulnerability in the FortiWeb web application firewall that allows a remote attacker to bypass authentication.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"research"},{"id":"e8d76cb5089f","published":"2025-08-16T12:30:32.000Z","title":"GHSA-v6cf-mv9h-c8mc / CVE-2025-9092: Bouncy Castle for Java Uncontrolled Resource Consumption Vulnerability","url":"https://github.com/advisories/GHSA-v6cf-mv9h-c8mc","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"bc062e73c9e4","published":"2025-08-16T10:41:00.004Z","title":"ERMAC V3.0 Banking Trojan Source Code Leak Exposes Full Malware Infrastructure","url":"https://thehackernews.com/2025/08/ermac-v30-banking-trojan-source-code.html","summary":"Cybersecurity researchers have detailed the inner workings of an Android banking trojan called ERMAC 3.0, uncovering serious shortcomings in the operators' infrastructure. \"The newly uncovered version 3.0 reveals a significant evolution of the malware, expanding its form…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"incident"},{"id":"45764d340e8b","published":"2025-08-16T05:34:00.002Z","title":"Russian Group EncryptHub Exploits MSC EvilTwin Vulnerability to Deploy Fickle Stealer Malware","url":"https://thehackernews.com/2025/08/russian-group-encrypthub-exploits-msc.html","summary":"The threat actor known as EncryptHub is continuing to exploit a now-patched security flaw impacting Microsoft Windows to deliver malicious payloads. Trustwave SpiderLabs said it recently observed an EncryptHub campaign that brings together social engineering and the…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"analysis"},{"id":"6f2b6aaf856e","published":"2025-08-16T01:51:59.000Z","title":"AI-Augmented CI/CD Pipelines: From Code Commit to Production with Autonomous Decisions","url":"https://arxiv.org/abs/2508.11867","summary":"Modern software delivery has accelerated from quarterly releases to multiple deployments per day. While CI/CD tooling has matured, human decision points interpreting flaky tests, choosing rollback strategies, tuning feature flags, and deciding when to promote a canary remain…","feed":"arxiv-agentic","publisher":"arxiv","topics":["trust"],"type":"research"},{"id":"86dbf247ea02","published":"2025-08-15T22:44:44.000Z","title":"The Summer of Johann: prompt injections as far as the eye can see","url":"https://simonwillison.net/2025/Aug/15/the-summer-of-johann","summary":"Independent AI researcher Johann Rehberger ( previously ) has had an absurdly busy August. Under the heading The Month of AI Bugs he has been publishing one report per day across an array of different tools, all of which are vulnerable to various classic prompt injection…","feed":"simonw","publisher":"simonw","topics":["agentic-sdlc","ai-security"],"type":"research"},{"id":"2fe5b6a014fc","published":"2025-08-15T22:13:54.000Z","title":"Rethinking Autonomy: Preventing Failures in AI-Driven Software Engineering","url":"https://arxiv.org/abs/2508.11824","summary":"The integration of Large Language Models (LLMs) into software engineering has revolutionized code generation, enabling unprecedented productivity through promptware and autonomous AI agents. However, this transformation introduces significant risks, including insecure code…","feed":"arxiv-agentic","publisher":"arxiv","topics":["trust"],"type":"research"},{"id":"3fb3351f4f64","published":"2025-08-15T21:31:18.000Z","title":"GHSA-wjrx-6529-hcj3 / CVE-2025-8959: HashiCorp go-getter Vulnerable to Symlink Attacks","url":"https://github.com/advisories/GHSA-wjrx-6529-hcj3","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"4e636666c145","published":"2025-08-15T20:34:20.000Z","title":"The Hallucinated Package Attack: Slopsquatting","url":"https://mend.io/blog/the-hallucinated-package-attack-slopsquatting","summary":"Learn how AI-generated code can lead to fake package installs and attacks.","feed":"mend","publisher":"mend","topics":[],"type":"analysis"},{"id":"79be48b4ab28","published":"2025-08-15T18:43:16.000Z","title":"GHSA-3x3q-ghcp-whf7 / CVE-2025-55285: Template Secret leakage in logs in Scaffolder when using `fetch:template`","url":"https://github.com/advisories/GHSA-3x3q-ghcp-whf7","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"e3ba8a7d1843","published":"2025-08-15T17:07:54.000Z","title":"CryptoScope: Utilizing Large Language Models for Automated Cryptographic Logic Vulnerability Detection","url":"https://arxiv.org/abs/2508.11599","summary":"Cryptographic algorithms are fundamental to modern security, yet their implementations frequently harbor subtle logic flaws that are hard to detect. We introduce CryptoScope, a novel framework for automated cryptographic vulnerability detection powered by Large Language Models…","feed":"arxiv-crse","publisher":"arxiv","topics":[],"type":"research"},{"id":"ca3041b5320f","published":"2025-08-15T16:52:48.000Z","title":"GHSA-9x9c-ghc5-jhw9 / CVE-2025-55207: @astrojs/node's trailing slash handling causes open redirect issue","url":"https://github.com/advisories/GHSA-9x9c-ghc5-jhw9","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"58f47116f002","published":"2025-08-15T15:25:00.000Z","title":"Colt Telecom attack claimed by WarLock ransomware, data up for sale","url":"https://bleepingcomputer.com/news/security/colt-telecom-attack-claimed-by-warlock-ransomware-data-up-for-sale","summary":"UK-based telecommunications company Colt Technology Services is dealing with a cyberattack that has caused a multi-day outage of some of the company's operations, including hosting and porting services, Colt Online and Voice API platforms.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"5f78f02f02e2","published":"2025-08-15T13:57:00.000Z","title":"Cisco warns of max severity flaw in Firewall Management Center","url":"https://bleepingcomputer.com/news/security/cisco-warns-of-max-severity-flaw-in-firewall-management-center","summary":"Cisco is warning about a critical remote code execution (RCE) vulnerability in the RADIUS subsystem of its Secure Firewall Management Center (FMC) software.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"157249c91d7d","published":"2025-08-15T11:57:23.000Z","title":"CoSAI Presentation at the 2025 All Things Open Conference","url":"https://coalitionforsecureai.org/cosai-presentation-at-the-2025-all-things-open-conference","summary":"CoSAI will be at the All Things Open Conference (12-14 Oct, Raleigh, NC). Andre Elizondo, Wiz, presents \"Hack-Proof AI: Building Security Through Collaboration\" on 13 Oct, 1:45-2:30PM ET, covering the CoSAI Workstreams.","feed":"cosai","publisher":"cosai","topics":["trust"],"type":"analysis"},{"id":"b90d5bc5fd07","published":"2025-08-15T11:42:07.000Z","title":"GHSA-m97p-h3c2-2rm5: Malware in meid-cli","url":"https://github.com/advisories/GHSA-m97p-h3c2-2rm5","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-malware","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a44c2482e2a2","published":"2025-08-15T11:41:00.000Z","title":"Plex warns users to patch security vulnerability immediately","url":"https://bleepingcomputer.com/news/security/plex-warns-users-to-patch-security-vulnerability-immediately","summary":"Plex has notified some of its users on Thursday to urgently update their media servers due to a recently patched security vulnerability.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"analysis"},{"id":"788d52287d7d","published":"2025-08-15T09:20:58.000Z","title":"Google Jules is Vulnerable To Invisible Prompt Injection","url":"https://embracethered.com/blog/posts/2025/google-jules-invisible-prompt-injection","summary":"The latest Gemini models quite reliably interpret hidden Unicode Tag characters as instructions . This vulnerability, first reported to Google over a year ago, has not been mitigated at the model or API level, hence now affects all applications built on top of Gemini. This…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["ai-security"],"type":"analysis"},{"id":"585f23ba7aec","published":"2025-08-15T06:49:00.000Z","title":"Cisco Warns of CVSS 10.0 FMC RADIUS Flaw Allowing Remote Code Execution","url":"https://thehackernews.com/2025/08/cisco-warns-of-cvss-100-fmc-radius-flaw.html","summary":"Cisco has released security updates to address a maximum-severity security flaw in Secure Firewall Management Center (FMC) Software that could allow an attacker to execute arbitrary code on affected systems. The vulnerability, assigned the CVE identifier CVE-2025-20265 (CVSS…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"analysis"},{"id":"d326e33b3644","published":"2025-08-15T05:03:26.000Z","title":"ORFuzz: Fuzzing the \"Other Side\" of LLM Safety -- Testing Over-Refusal","url":"https://arxiv.org/abs/2508.11222","summary":"Large Language Models (LLMs) increasingly exhibit over-refusal - erroneously rejecting benign queries due to overly conservative safety measures - a critical functional flaw that undermines their reliability and usability. Current methods for testing this behavior are…","feed":"arxiv-dast","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"2ddca1489c2e","published":"2025-08-15T00:00:00.000Z","title":"Should Security Solutions Be Secure? Maybe We're All Wrong - Fortinet FortiSIEM Pre-Auth Command Injection (CVE-2025-25256)","url":"https://labs.watchtowr.com/should-security-solutions-be-secure-maybe-were-all-wrong-fortinet-fortisiem-pre-auth-command-injection-cve-2025-25256","summary":"","feed":"watchtowr","publisher":"watchtowr","topics":[],"type":"analysis"},{"id":"8111bfd6668b","published":"2025-08-15T00:00:00.000Z","title":"Announcing the Sigstore Transparency Log Research Dataset","url":"https://blog.sigstore.dev/rekor-bigquery-dataset","summary":"We’re pleased to announce the creation of a new BigQuery public dataset, rekor. The rekor dataset is an easily-queryable mirror of the public good instance of Sigstore’s transparency log, Rekor. As a reminder, signing events are recorded in Rekor, Sigstore’s append-only…","feed":"sigstore","publisher":"sigstore","topics":["trust"],"type":"analysis"},{"id":"0d797f2917c7","published":"2025-08-14T23:06:39.000Z","title":"Risky Biz Soap Box: How to measure vulnerability reachability","url":"https://risky.biz/soapbox99","summary":"In this Soap Box edition of the Risky Business podcast Patrick Gray chats with Socket founder Feross Aboukhadijeh about how to measure the reachability of vulnerabilities in applications. It’s great to know there’s a CVE in a library you’re using, but it’s even better if you…","feed":"riskybiz","publisher":"riskybiz","topics":[],"type":"guide"},{"id":"de539b445441","published":"2025-08-14T22:23:02.000Z","title":"GHSA-77h3-w9rx-hj3q: User-defined implementations of the safe trait scratchpad::Tracking can cause heap buffer overflows","url":"https://github.com/advisories/GHSA-77h3-w9rx-hj3q","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d14d052a2870","published":"2025-08-14T20:22:52.000Z","title":"Case Study: How LFX Insights and OSPS Baseline Validated GUAC’s Security in Under an Hour","url":"https://openssf.org/blog/2025/08/14/case-study-how-lfx-insights-and-osps-baseline-validated-guacs-security-in-under-an-hour","summary":"Tools: GUAC, OSPS Baseline, LFX Insights Challenge: Demonstrating strong security posture quickly and credibly to stakeholders Solution: Leveraging Linux Foundation Insights (LFX Insights) and the Open Source Security Foundation (OpenSSF) Open Source Project Security Baseline…","feed":"openssf","publisher":"openssf","topics":["sca"],"type":"research"},{"id":"703f06f0073b","published":"2025-08-14T20:01:33.000Z","title":"Show HN: MCP Security Suite","url":"https://github.com/NineSunsInc/mighty-security","summary":"Hacker News: 36 points, 32 comments.","feed":"hn-ai-security-0","publisher":"hacker-news","topics":["ai-security"],"type":"analysis"},{"id":"95c1e9e6c65d","published":"2025-08-14T18:39:04.000Z","title":"CoSAI at Cyber EUnnovate 2025","url":"https://coalitionforsecureai.org/cosai-at-cyber-eunnovate-2025","summary":"CoSAI will be represented at Cyber EUnnovate 2025, a premier conference exploring the intersection of AI innovation and cybersecurity across Europe.","feed":"cosai","publisher":"cosai","topics":["trust"],"type":"analysis"},{"id":"089d9b26429f","published":"2025-08-14T18:30:08.000Z","title":"GHSA-wm7x-ww72-r77q / CVE-2025-9039: Information Disclosure in Amazon ECS Container Agent","url":"https://github.com/advisories/GHSA-wm7x-ww72-r77q","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"982109c4ae2f","published":"2025-08-14T18:00:25.000Z","title":"MCP-Guard: A Multi-Stage Defense-in-Depth Framework for Securing Model Context Protocol in Agentic AI","url":"https://arxiv.org/abs/2508.10991","summary":"While Large Language Models (LLMs) have achieved remarkable performance, they remain vulnerable to jailbreak. The integration of Large Language Models (LLMs) with external tools via protocols such as the Model Context Protocol (MCP) introduces critical security vulnerabilities,…","feed":"arxiv-agentic","publisher":"arxiv","topics":["sast","models","ai-security"],"type":"research"},{"id":"5ca29c92c268","published":"2025-08-14T17:49:09.000Z","title":"Searching for Privacy Risks in LLM Agents via Simulation","url":"https://arxiv.org/abs/2508.10880","summary":"The widespread deployment of LLM-based agents is likely to introduce a critical privacy threat: malicious agents that proactively engage others in multi-turn interactions to extract sensitive information. However, the evolving nature of such dynamic dialogues makes it…","feed":"arxiv-agentic","publisher":"arxiv","topics":[],"type":"research"},{"id":"71560db07d25","published":"2025-08-14T17:14:16.000Z","title":"OpenSSF at Black Hat USA 2025 & DEF CON 33: AIxCC Highlights, Big Wins, and the Future of Securing Open Source","url":"https://openssf.org/blog/2025/08/14/openssf-at-black-hat-usa-2025-def-con-33-aixcc-highlights-big-wins-and-the-future-of-securing-open-source","summary":"The Open Source Security Foundation (OpenSSF) marked a strong presence at two cornerstone cybersecurity events, Black Hat USA 2025 and DEF CON 33, engaging with security leaders, showcasing our initiatives, and fostering collaboration to advance open source security.","feed":"openssf","publisher":"openssf","topics":[],"type":"analysis"},{"id":"d5e8aa60c4dd","published":"2025-08-14T16:43:35.000Z","title":"Benchmark Dataset Generation and Evaluation for Excel Formula Repair with LLMs","url":"https://arxiv.org/abs/2508.11715","summary":"Excel is a pervasive yet often complex tool, particularly for novice users, where runtime errors arising from logical mistakes or misinterpretations of functions pose a significant challenge. While large language models (LLMs) offer promising assistance by explaining formula…","feed":"arxiv-agentic","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"5b4e728dc500","published":"2025-08-14T16:39:04.000Z","title":"GHSA-j26p-6wx7-f3pw / CVE-2025-54867: Youki: If /proc and /sys in the rootfs are symbolic links, they can potentially be exploited to gain access to the host root filesystem.","url":"https://github.com/advisories/GHSA-j26p-6wx7-f3pw","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"2504c18f1cf6","published":"2025-08-14T15:30:44.000Z","title":"GHSA-9g5x-mm39-wg9r / CVE-2025-55673: Apache Superset data query improperly discloses database schema information to low-privileged guest user","url":"https://github.com/advisories/GHSA-9g5x-mm39-wg9r","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"27db7e1edc0a","published":"2025-08-14T15:30:44.000Z","title":"GHSA-mhpq-m962-mg92 / CVE-2025-55675: Apache Superset allows authenticated users to discover metadata about datasources they don't have permission to access","url":"https://github.com/advisories/GHSA-mhpq-m962-mg92","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"9232b538910d","published":"2025-08-14T15:30:44.000Z","title":"GHSA-fxgf-3xh6-m2pp / CVE-2025-55674: Apache Superset has bypass of `DISALLOWED_SQL_FUNCTIONS` that allows execution of blocked SQL functions","url":"https://github.com/advisories/GHSA-fxgf-3xh6-m2pp","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"9cb5ba02a02b","published":"2025-08-14T15:30:44.000Z","title":"GHSA-fj97-2v9x-w5m4 / CVE-2025-55672: Apache Superset's chart visualization has a stored Cross-Site Scripting (XSS) vulnerability","url":"https://github.com/advisories/GHSA-fj97-2v9x-w5m4","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"738e64a5ca66","published":"2025-08-14T15:30:22.000Z","title":"Data and Context Matter: Towards Generalizing AI-based Software Vulnerability Detection","url":"https://arxiv.org/abs/2508.16625","summary":"AI-based solutions demonstrate remarkable results in identifying vulnerabilities in software, but research has consistently found that this performance does not generalize to unseen codebases. In this paper, we specifically investigate the impact of model architecture,…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"69b786eee769","published":"2025-08-14T15:20:00.008Z","title":"New HTTP/2 'MadeYouReset' Vulnerability Enables Large-Scale DoS Attacks","url":"https://thehackernews.com/2025/08/new-http2-madeyoureset-vulnerability.html","summary":"Multiple HTTP/2 implementations have been found susceptible to a new attack technique called MadeYouReset that could be explored to conduct powerful denial-of-service (DoS) attacks. \"MadeYouReset bypasses the typical server-imposed limit of 100 concurrent HTTP/2 requests per…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"79086a140412","published":"2025-08-14T15:05:00.000Z","title":"Microsoft fixes Windows Server bug causing cluster, VM issues","url":"https://bleepingcomputer.com/news/microsoft/microsoft-fixes-windows-server-bug-causing-cluster-vm-issues","summary":"Microsoft has resolved a known issue that triggers Cluster service and VM restart issues after installing July's Windows Server 2019 security updates.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"analysis"},{"id":"69798a41280a","published":"2025-08-14T12:30:23.000Z","title":"GHSA-2vv2-3x8x-4gv7 / CVE-2025-8943: Flowise OS command remote code execution","url":"https://github.com/advisories/GHSA-2vv2-3x8x-4gv7","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"12d7a2698147","published":"2025-08-14T11:20:58.000Z","title":"Jules Zombie Agent: From Prompt Injection to Remote Control","url":"https://embracethered.com/blog/posts/2025/google-jules-remote-code-execution-zombai","summary":"In the previous post , we explored two data exfiltration vectors that Jules is vulnerable to and that can be exploited via prompt injection. This post takes it further by demonstrating how Jules can be convinced to download malware and join a remote command & control server.…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["ai-security"],"type":"analysis"},{"id":"5ac88058b8af","published":"2025-08-14T11:08:52.000Z","title":"LLM Coding Integrity Breach","url":"https://schneier.com/blog/archives/2025/08/llm-coding-integrity-breach.html","summary":"Here’s an interesting story about a failure being introduced by LLM-written code. Specifically, the LLM was doing some code refactoring, and when it moved a chunk of code from one file to another it changed a “break” to a “continue.” That turned an error logging statement into…","feed":"schneier","publisher":"schneier","topics":[],"type":"incident"},{"id":"fc440e98fa7e","published":"2025-08-14T10:45:00.000Z","title":"Leak: OpenAI's browser will use ChatGPT Agent to control the browser","url":"https://bleepingcomputer.com/news/artificial-intelligence/leak-openais-browser-will-use-chatgpt-agent-to-control-the-browser","summary":"OpenAI is building an agentic future with its upcoming Chromium-based browser and a new leak confirms GPT Agent integration.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"e7e7376a2df3","published":"2025-08-14T08:30:44.000Z","title":"Just Released: “The Future of AppSec in the Era of AI” 2026 Industry Outlook","url":"https://checkmarx.com/blog/ai-llm-tools-in-application-security/just-released-the-future-of-appsec-in-the-era-of-ai-2026-industry-outlook","summary":"When a False Sense of Security Meets Breakneck Developer Velocity In an industry long accustomed to rapid change, 2025 may be remembered as the year the ground truly shifted. According to The Future of Application Security Report, a global research study conducted by Censuswide…","feed":"checkmarx","publisher":"checkmarx","topics":[],"type":"analysis"},{"id":"0f36ab3986aa","published":"2025-08-14T07:35:11.000Z","title":"Enhancing GraphQL Security by Detecting Malicious Queries Using Large Language Models, Sentence Transformers, and Convolutional Neural Networks","url":"https://arxiv.org/abs/2508.11711","summary":"GraphQL's flexibility, while beneficial for efficient data fetching, introduces unique security vulnerabilities that traditional API security mechanisms often fail to address. Malicious GraphQL queries can exploit the language's dynamic nature, leading to denial-of-service…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"84e3e916393c","published":"2025-08-14T06:46:56.000Z","title":"Jailbreaking Commercial Black-Box LLMs with Explicitly Harmful Prompts","url":"https://arxiv.org/abs/2508.10390","summary":"Existing black-box jailbreak attacks achieve certain success on non-reasoning models but degrade significantly on recent SOTA reasoning models. To improve attack ability, inspired by adversarial aggregation strategies, we integrate multiple jailbreak tricks into a single…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"048489f62ce4","published":"2025-08-14T05:41:58.000Z","title":"Code Vulnerability Detection Across Different Programming Languages with AI Models","url":"https://arxiv.org/abs/2508.11710","summary":"Security vulnerabilities present in a code that has been written in diverse programming languages are among the most critical yet complicated aspects of source code to detect. Static analysis tools based on rule-based patterns usually do not work well at detecting the…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"3c3d3e66c4cd","published":"2025-08-14T03:39:54.000Z","title":"A Vision-Language Pre-training Model-Guided Approach for Mitigating Backdoor Attacks in Federated Learning","url":"https://arxiv.org/abs/2508.10315","summary":"Defending backdoor attacks in Federated Learning (FL) under heterogeneous client data distributions encounters limitations balancing effectiveness and privacy-preserving, while most existing methods highly rely on the assumption of homogeneous client data distributions or the…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"72d3e71bd68e","published":"2025-08-14T01:20:58.000Z","title":"Google Jules: Vulnerable to Multiple Data Exfiltration Issues","url":"https://embracethered.com/blog/posts/2025/google-jules-vulnerable-to-data-exfiltration-issues","summary":"This post explores data exfiltration attacks in Google Jules, an asynchronous coding agent. This is the first of three posts that will highlight my research on Google Jules in May 2025. All information provided was also shared with Google at that time. This first post will…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc"],"type":"analysis"},{"id":"10edea3488cf","published":"2025-08-14T00:13:22.000Z","title":"Pruning and Malicious Injection: A Retraining-Free Backdoor Attack on Transformer Models","url":"https://arxiv.org/abs/2508.10243","summary":"Transformer models have demonstrated exceptional performance and have become indispensable in computer vision (CV) and natural language processing (NLP) tasks. However, recent studies reveal that transformers are susceptible to backdoor attacks. Prior backdoor attack methods…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"77de855559e0","published":"2025-08-14T00:06:00.000Z","title":"GHSA-r4mg-4433-c7g3 / CVE-2025-24293: Active Storage allowed transformation methods that were potentially unsafe","url":"https://github.com/advisories/GHSA-r4mg-4433-c7g3","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"07782a20cec9","published":"2025-08-14T00:05:27.000Z","title":"GHSA-f9f8-9pmf-xv68 / CVE-2025-55198: Helm May Panic Due To Incorrect YAML Content","url":"https://github.com/advisories/GHSA-f9f8-9pmf-xv68","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d07ebaeec1d7","published":"2025-08-14T00:01:34.000Z","title":"GHSA-9h84-qmv7-982p / CVE-2025-55199: Helm Charts with Specific JSON Schema Values Can Cause Memory Exhaustion","url":"https://github.com/advisories/GHSA-9h84-qmv7-982p","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"2d28aefee31a","published":"2025-08-14T00:00:00.000Z","title":"Announcing Kernel-Independent FIPS for Java","url":"https://chainguard.dev/unchained/announcing-kernel-independent-fips-for-java","summary":"Kernel-Independent FIPS is now available across the full catalog of Chainguard FIPS images for Java, simplifying and accelerating compliance for FedRAMP ATO.","feed":"chainguard","publisher":"chainguard","topics":["trust"],"type":"analysis"},{"id":"56ef65d1ddd0","published":"2025-08-13T23:54:02.000Z","title":"GHSA-xvr7-p2c6-j83w: swift-nio-http2 affected by HTTP/2 MadeYouReset vulnerability","url":"https://github.com/advisories/GHSA-xvr7-p2c6-j83w","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b24c94a06d2f","published":"2025-08-13T23:03:26.000Z","title":"GHSA-x6gv-2rvh-qmp6: m00nl1ght-dev/steam-workshop-deploy: Exposure of Version-Control Repository to an Unauthorized Control Sphere and Insufficiently Protected Credentials","url":"https://github.com/advisories/GHSA-x6gv-2rvh-qmp6","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"50dd33ac6268","published":"2025-08-13T22:32:18.000Z","title":"GHSA-76r7-hhxj-r776 / CVE-2025-55193: Active Record logging vulnerable to ANSI escape injection","url":"https://github.com/advisories/GHSA-76r7-hhxj-r776","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a974ce519103","published":"2025-08-13T22:05:07.000Z","title":"Creating AI agent solutions for warehouse data access and security","url":"https://engineering.fb.com/2025/08/13/data-infrastructure/agentic-solution-for-warehouse-data-access","summary":"In this post, we explore the ways we’re evolving Meta’s data warehouse to facilitate productivity and security to serve both human users and AI agents. We detail how we’re developing agents that help users making data access requests to get to the data they need, and that help…","feed":"meta-eng","publisher":"meta-eng","topics":["ai-security"],"type":"research"},{"id":"5d3e25112b83","published":"2025-08-13T21:30:30.000Z","title":"GHSA-97vm-c39p-jr86 / CVE-2011-10019: Spree has Remote Command Execution vulnerability in search functionality","url":"https://github.com/advisories/GHSA-97vm-c39p-jr86","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"94660047b0d2","published":"2025-08-13T20:23:00.000Z","title":"Fortinet warns of FortiSIEM pre-auth RCE flaw with exploit in the wild","url":"https://bleepingcomputer.com/news/security/fortinet-warns-of-fortisiem-pre-auth-rce-flaw-with-exploit-in-the-wild","summary":"Fortinet is warning about a remote unauthenticated command injection flaw in FortiSIEM that has in-the-wild exploit code, making it critical for admins to apply the latest security updates.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"analysis"},{"id":"9f30a2d7d6db","published":"2025-08-13T19:51:24.000Z","title":"GHSA-7hfw-26vp-jp8m / CVE-2025-55197: PyPDF's Manipulated FlateDecode streams can exhaust RAM","url":"https://github.com/advisories/GHSA-7hfw-26vp-jp8m","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"86cc77869bc1","published":"2025-08-13T19:45:15.000Z","title":"GHSA-fcxq-v2r3-cc8h / CVE-2025-55196: External Secrets Operator's Missing Namespace Restriction Allows Unauthorized Secret Access","url":"https://github.com/advisories/GHSA-fcxq-v2r3-cc8h","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"742ed2511a7f","published":"2025-08-13T19:24:58.000Z","title":"CoSAI to Participate in Justice, Public Safety, and Security Artificial Intelligence (AI) Summit","url":"https://coalitionforsecureai.org/cosai-to-participate-in-justice-public-safety-and-security-artificial-intelligence-ai-summit","summary":"CoSAI will be represented at the Justice, Public Safety, and Security Artificial Intelligence (AI) Summit, with a panel on AI Security and Standards on 10 September.","feed":"cosai","publisher":"cosai","topics":[],"type":"analysis"},{"id":"cf8feb95424e","published":"2025-08-13T19:21:35.000Z","title":"Fortinet SIEM issue coincides with spike in brute-force traffic against company’s SSL VPNs","url":"https://cyberscoop.com/fortinet-fortisiem-critical-vulnerability-ssl-vpn-brute-force-traffic","summary":"Researchers aren’t aware of any active exploitation of the software, but the issue is being dealt with simultaneously as attackers are trying to brute force the company’s security appliances.","feed":"cyberscoop","publisher":"cyberscoop","topics":[],"type":"research"},{"id":"20052d09e702","published":"2025-08-13T19:06:56.000Z","title":"GHSA-prj3-ccx8-p6x4 / CVE-2025-55163: Netty affected by MadeYouReset HTTP/2 DDoS vulnerability","url":"https://github.com/advisories/GHSA-prj3-ccx8-p6x4","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b23897de8428","published":"2025-08-13T18:47:35.000Z","title":"GHSA-gpmg-4x4g-mr5r / CVE-2025-54791: OMERO.web displays unecessary user information when requesting password reset","url":"https://github.com/advisories/GHSA-gpmg-4x4g-mr5r","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"587501733851","published":"2025-08-13T18:31:25.000Z","title":"GHSA-p3qf-84rg-jxfc / CVE-2025-50946: OliveTin OS Command Injection vulnerability","url":"https://github.com/advisories/GHSA-p3qf-84rg-jxfc","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"12d2100e514a","published":"2025-08-13T17:26:28.000Z","title":"177 AWS services achieve HITRUST certification","url":"https://aws.amazon.com/blogs/security/177-aws-services-achieve-hitrust-certification","summary":"Amazon Web Services (AWS) is excited to announce that 177 AWS services have achieved HITRUST certification for the 2025 assessment cycle, including the following five services which were certified for the first time: Amazon Verified Permissions AWS B2B Data Interchange AWS…","feed":"aws-security","publisher":"aws-security","topics":["trust"],"type":"analysis"},{"id":"9c6dad0d3b9a","published":"2025-08-13T16:42:00.000Z","title":"Spike in Fortinet VPN brute-force attacks raises zero-day concerns","url":"https://bleepingcomputer.com/news/security/spike-in-fortinet-vpn-brute-force-attacks-raises-zero-day-concerns","summary":"A massive spike in brute-force attacks targeted Fortinet SSL VPNs earlier this month, followed by a switch to FortiManager, marked a deliberate shift in targeting that has historically preceded new vulnerability disclosures.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"0a621022f6d4","published":"2025-08-13T15:46:00.002Z","title":"New PS1Bot Malware Campaign Uses Malvertising to Deploy Multi-Stage In-Memory Attacks","url":"https://thehackernews.com/2025/08/new-ps1bot-malware-campaign-uses.html","summary":"Cybersecurity researchers have discovered a new malvertising campaign that's designed to infect victims with a multi-stage malware framework called PS1Bot. \"PS1Bot features a modular design, with several modules delivered used to perform a variety of malicious activities on…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"incident"},{"id":"62d367ffbdd2","published":"2025-08-13T15:30:34.000Z","title":"GHSA-gqp3-2cvr-x8m3 / CVE-2025-48989: Apache Tomcat Improper Resource Shutdown or Release vulnerability","url":"https://github.com/advisories/GHSA-gqp3-2cvr-x8m3","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"707d55a8b2bc","published":"2025-08-13T15:30:34.000Z","title":"GHSA-vq9x-w82r-rhmc / CVE-2025-52392: Soosyze CMS's /user/login endpoint missing rate-limiting and lockout mechanisms","url":"https://github.com/advisories/GHSA-vq9x-w82r-rhmc","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"fd2d99f8e7c6","published":"2025-08-13T15:30:34.000Z","title":"GHSA-23hv-mwm6-g8jf / CVE-2025-55668: Apache Tomcat Session Fixation vulnerability","url":"https://github.com/advisories/GHSA-23hv-mwm6-g8jf","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a1af00a038e7","published":"2025-08-13T14:00:00.000Z","title":"How we found TeaOnHer spilling users’ driver’s licenses in less than 10 minutes","url":"https://techcrunch.com/2025/08/13/how-we-found-teaonher-spilling-users-drivers-licenses-in-less-than-10-minutes","summary":"Exclusive: A dating gossip app for men exposed thousands of users' personal data, including scans of driver's licenses. The app's developer, Xavier Lampkin, won't say if he plans to notify affected users about the app's security lapse.","feed":"techcrunch-security","publisher":"techcrunch-security","topics":[],"type":"research"},{"id":"2ebb8935a1f5","published":"2025-08-13T13:47:55.000Z","title":"Extending the OWASP Multi-Agentic System Threat Modeling Guide: Insights from Multi-Agent Security Research","url":"https://arxiv.org/abs/2508.09815","summary":"We propose an extension to the OWASP Multi-Agentic System (MAS) Threat Modeling Guide, translating recent anticipatory research in multi-agent security (MASEC) into practical guidance for addressing challenges unique to large language model (LLM)-driven multi-agent…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"f5cb1fb1f8d2","published":"2025-08-13T13:24:39.000Z","title":"Patch the vulnerability: Confirm Sean Plankey as CISA director","url":"https://cyberscoop.com/sean-plankey-cisa-confirmation-op-ed","summary":"The executive director of the National Technology Security Coalition writes that Plankey is a strong, capable leader who will strengthen public-private partnerships.","feed":"cyberscoop","publisher":"cyberscoop","topics":["remediation"],"type":"analysis"},{"id":"34f1f59ad2ca","published":"2025-08-13T13:19:00.003Z","title":"Zoom and Xerox Release Critical Security Updates Fixing Privilege Escalation and RCE Flaws","url":"https://thehackernews.com/2025/08/zoom-and-xerox-release-critical.html","summary":"Zoom and Xerox have addressed critical security flaws in Zoom Clients for Windows and FreeFlow Core that could allow privilege escalation and remote code execution. The vulnerability impacting Zoom Clients for Windows, tracked as CVE-2025-49457 (CVSS score: 9.6), relates to a…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"release"},{"id":"f147cbea97d2","published":"2025-08-13T12:31:30.000Z","title":"GHSA-4cx2-fc23-5wg6 / CVE-2025-8916: Bouncy Castle for Java bcpkix, bcprov, bcpkix-fips on All (API modules) allows Excessive Allocation","url":"https://github.com/advisories/GHSA-4cx2-fc23-5wg6","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a987bfbc49fd","published":"2025-08-13T11:37:00.004Z","title":"Fortinet Warns About FortiSIEM Vulnerability (CVE-2025-25256) With In-the-Wild Exploit Code","url":"https://thehackernews.com/2025/08/fortinet-warns-about-fortisiem.html","summary":"Fortinet is alerting customers of a critical security flaw in FortiSIEM for which it said there exists an exploit in the wild. The vulnerability, tracked as CVE-2025-25256, carries a CVSS score of 9.8 out of a maximum of 10.0. \"An improper neutralization of special elements…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"84ef873ec7b2","published":"2025-08-13T11:25:00.001Z","title":"AI SOC 101: Key Capabilities Security Leaders Need to Know","url":"https://thehackernews.com/2025/08/ai-soc-101-key-capabilities-security.html","summary":"Security operations have never been a 9-to-5 job. For SOC analysts, the day often starts and ends deep in a queue of alerts, chasing down what turns out to be false positives, or switching between half a dozen tools to piece together context. The work is repetitive,…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"73480d4ffa97","published":"2025-08-13T09:35:51.000Z","title":"Demystifying the Role of Rule-based Detection in AI Systems for Windows Malware Detection","url":"https://arxiv.org/abs/2508.09652","summary":"Malware detection increasingly relies on AI systems that integrate signature-based detection with machine learning. However, these components are typically developed and combined in isolation, missing opportunities to reduce data complexity and strengthen defenses against…","feed":"arxiv-dast","publisher":"arxiv","topics":[],"type":"research"},{"id":"e80b97617ee6","published":"2025-08-13T08:47:00.009Z","title":"Microsoft August 2025 Patch Tuesday Fixes Kerberos Zero-Day Among 111 Total New Flaws","url":"https://thehackernews.com/2025/08/microsoft-august-2025-patch-tuesday.html","summary":"Microsoft on Tuesday rolled out fixes for a massive set of 111 security flaws across its software portfolio, including one flaw that has been disclosed as publicly known at the time of the release. Of the 111 vulnerabilities, 16 are rated Critical, 92 are rated Important, two…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"advisory"},{"id":"1966bb3bb6b8","published":"2025-08-13T00:00:00.000Z","title":"CVE-2025-8875: N-able N-Central Insecure Deserialization Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2025-8875","summary":"N-able N-Central contains an insecure deserialization vulnerability that could lead to command execution. US federal agencies must act by 2025-08-20.","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"f0a17e09e098","published":"2025-08-13T00:00:00.000Z","title":"CVE-2025-8876: N-able N-Central Command Injection Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2025-8876","summary":"N-able N-Central contains a command injection vulnerability via improper sanitization of user input. US federal agencies must act by 2025-08-20.","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"f9b712edf1f3","published":"2025-08-12T23:46:30.000Z","title":"Reclaiming Constitutional Authority of Algorithmic Power","url":"https://arxiv.org/abs/2508.11699","summary":"Whether and how to govern AI is no longer a question of technical regulation. It is a question of constitutional authority. Across jurisdictions, algorithmic systems now perform functions once reserved to public institutions: allocating welfare, determining legal status,…","feed":"arxiv-trust","publisher":"arxiv","topics":["trust"],"type":"research"},{"id":"1feb12c980a3","published":"2025-08-12T22:14:41.000Z","title":"Microsoft Patch Tuesday, August 2025 Edition","url":"https://krebsonsecurity.com/2025/08/microsoft-patch-tuesday-august-2025-edition","summary":"Microsoft today released updates to fix more than 100 security flaws in its Windows operating systems and other software. At least 13 of the bugs received Microsoft's most-dire \"critical\" rating, meaning they could be abused by malware or malcontents to gain remote access to a…","feed":"krebs","publisher":"krebs","topics":[],"type":"roundup"},{"id":"da14dc2f1fe4","published":"2025-08-12T21:31:20.000Z","title":"GHSA-m5c7-5gv3-hcpf / CVE-2025-43734: Liferay Portal 7.4.0 and Liferay DXP have a reflected cross-site scripting (XSS) vulnerability","url":"https://github.com/advisories/GHSA-m5c7-5gv3-hcpf","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b172c45327ed","published":"2025-08-12T21:20:58.000Z","title":"GitHub Copilot: Remote Code Execution via Prompt Injection (CVE-2025-53773)","url":"https://embracethered.com/blog/posts/2025/github-copilot-remote-code-execution-via-prompt-injection","summary":"This post is about an important, but also scary, prompt injection discovery that leads to full system compromise of the developer’s machine in GitHub Copilot and VS Code . It is achieved by placing Copilot into YOLO mode by modifying the project’s settings.json file. As…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","ai-security"],"type":"analysis"},{"id":"0411e03aed3f","published":"2025-08-12T20:21:46.000Z","title":"Microsoft Patch Tuesday follows SharePoint attacks, Exchange server warnings","url":"https://cyberscoop.com/microsoft-patch-tuesday-august-2025","summary":"Despite serious alarm raised by officials, organizations have not applied the patch for Microsoft Exchange servers en masse.","feed":"cyberscoop","publisher":"cyberscoop","topics":[],"type":"analysis"},{"id":"29e088732707","published":"2025-08-12T20:20:58.000Z","title":"GHSA-22wq-q86m-83fh / CVE-2025-55166: svg-sanitizer Bypasses Attribute Sanitization","url":"https://github.com/advisories/GHSA-22wq-q86m-83fh","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"90396c74b7b9","published":"2025-08-12T20:03:03.000Z","title":"Securing AI: The Next Cybersecurity Battleground","url":"https://openssf.org/blog/2025/08/12/securing-ai-the-next-cybersecurity-battleground","summary":"The AI wave is here, and it’s only getting bigger. According to a recent report from McKinsey, “over the next three years, 92 percent of companies plan to increase their AI investments.” As this AI wave washes over almost every industry and is integrated deeply and extensively…","feed":"openssf","publisher":"openssf","topics":["models","ai-security"],"type":"analysis"},{"id":"981e3b184b2b","published":"2025-08-12T19:33:07.000Z","title":"GHSA-c9rc-mg46-23w3 / CVE-2025-8747: Keras vulnerable to CVE-2025-1550 bypass via reuse of internal functionality","url":"https://github.com/advisories/GHSA-c9rc-mg46-23w3","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"13c5b9c67877","published":"2025-08-12T18:47:30.000Z","title":"Can AI Keep a Secret? Contextual Integrity Verification: A Provable Security Architecture for LLMs","url":"https://arxiv.org/abs/2508.09288","summary":"Large language models (LLMs) remain acutely vulnerable to prompt injection and related jailbreak attacks; heuristic guardrails (rules, filters, LLM judges) are routinely bypassed. We present Contextual Integrity Verification (CIV), an inference-time security architecture that…","feed":"arxiv-models","publisher":"arxiv","topics":["models","ai-security"],"type":"research"},{"id":"326c5576de85","published":"2025-08-12T18:43:00.000Z","title":"Docker Hub still hosts dozens of Linux images with the XZ backdoor","url":"https://bleepingcomputer.com/news/security/docker-hub-still-hosts-dozens-of-linux-images-with-the-xz-backdoor","summary":"The XZ-Utils backdoor, first discovered in March 2024, is still present in at least 35 Linux images on Docker Hub, potentially putting users, organizations, and their data at risk.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["sca"],"type":"analysis"},{"id":"25cb2a874fa1","published":"2025-08-12T18:31:30.000Z","title":"GHSA-5777-jj7p-mpqw / CVE-2025-49555: Magento Cross-Site Request Forgery (CSRF) vulnerability","url":"https://github.com/advisories/GHSA-5777-jj7p-mpqw","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6c2a3e6a3353","published":"2025-08-12T18:31:30.000Z","title":"GHSA-xgfm-992v-h2hr / CVE-2025-49554: Magento vulnerable to denial of service","url":"https://github.com/advisories/GHSA-xgfm-992v-h2hr","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"84151c5b8b49","published":"2025-08-12T18:31:30.000Z","title":"GHSA-7hrj-3c9x-xv5h / CVE-2025-49556: Magento has incorrect authorization issue that leads to arbitrary file system read","url":"https://github.com/advisories/GHSA-7hrj-3c9x-xv5h","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b653048470b3","published":"2025-08-12T18:31:30.000Z","title":"GHSA-8mq8-c243-2335 / CVE-2025-49557: Magento Cross-site Scripting vulnerability","url":"https://github.com/advisories/GHSA-8mq8-c243-2335","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b896c1c64f9a","published":"2025-08-12T18:31:30.000Z","title":"GHSA-h4f4-gv6h-x824 / CVE-2025-49559: Magento vulnerable to path traversal","url":"https://github.com/advisories/GHSA-h4f4-gv6h-x824","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d402db623606","published":"2025-08-12T18:31:30.000Z","title":"GHSA-wcmw-8xpp-rwfj / CVE-2025-49558: Magento Time-of-check Time-of-use (TOCTOU) Race Condition vulnerability","url":"https://github.com/advisories/GHSA-wcmw-8xpp-rwfj","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"082d699aa9bb","published":"2025-08-12T18:17:00.002Z","title":"Researchers Spot XZ Utils Backdoor in Dozens of Docker Hub Images, Fueling Supply Chain Risks","url":"https://thehackernews.com/2025/08/researchers-spot-xz-utils-backdoor-in.html","summary":"New research has uncovered Docker images on Docker Hub that contain the infamous XZ Utils backdoor, more than a year after the discovery of the incident. More troubling is the fact that other images have been built on top of these infected base images, effectively propagating…","feed":"thehackernews","publisher":"thehackernews","topics":["sca"],"type":"research"},{"id":"5a23a742db7f","published":"2025-08-12T18:07:44.000Z","title":"GHSA-w2cq-g8g3-gm83 / CVE-2025-55164: content-security-policy-parser Prototype Pollution Vulnerability May Lead to RCE","url":"https://github.com/advisories/GHSA-w2cq-g8g3-gm83","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"7b94f1dedcb5","published":"2025-08-12T17:43:00.000Z","title":"Microsoft August 2025 Patch Tuesday fixes one zero-day, 107 flaws","url":"https://bleepingcomputer.com/news/microsoft/microsoft-august-2025-patch-tuesday-fixes-one-zero-day-107-flaws","summary":"Today is Microsoft's August 2025 Patch Tuesday, which includes security updates for 107 flaws, including one publicly disclosed zero-day vulnerability in Windows Kerberos.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"advisory"},{"id":"9e20f9ae1161","published":"2025-08-12T17:34:00.000Z","title":"Windows 11 KB5063878 & KB5063875 cumulative updates released","url":"https://bleepingcomputer.com/news/microsoft/windows-11-kb5063878-and-kb5063875-cumulative-updates-released","summary":"Microsoft has released Windows 11 KB5063878 and KB5063875 cumulative updates for versions 24H2 and 23H2 to fix security vulnerabilities and issues.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"analysis"},{"id":"2202981584a7","published":"2025-08-12T17:15:00.000Z","title":"Windows 10 KB5063709 update fixes extended security updates enrollment","url":"https://bleepingcomputer.com/news/microsoft/windows-10-kb5063709-update-fixes-extended-security-updates-enrollment","summary":"Microsoft has released the KB5063709 cumulative update for Windows 10 22H2 and Windows 10 21H2, with seven fixes or changes, including a fix for a bug that prevented enrollment in extended security updates.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"analysis"},{"id":"989421e82376","published":"2025-08-12T16:37:03.000Z","title":"What’s in the SOSS? Podcast #37 – S2E14 Open Source Security: OSTIF’s 10-Year Journey of Collaborative Audits","url":"https://openssf.org/podcast/2025/08/12/whats-in-the-soss-podcast-37-s2e14-open-source-security-ostifs-10-year-journey-of-collaborative-audits","summary":"Summary In this episode of “What’s in the SOSS,” Derek Zimmer and Amir Montezari from the Open Source Technology Improvement Fund (OSTIF) discuss their decade-long mission of providing security resources to open source projects. They focus on collaborative, maintainer-centric…","feed":"openssf","publisher":"openssf","topics":[],"type":"roundup"},{"id":"f75da727215c","published":"2025-08-12T16:22:46.000Z","title":"Dependabot version updates now support vcpkg","url":"https://github.blog/changelog/2025-08-12-dependabot-version-updates-now-support-vcpkg","summary":"Dependabot now supports automatic dependency updates for vcpkg, the free C/C++ package manager from Microsoft. This enables teams using vcpkg to keep their C and C++ dependencies secure and up-to-date…","feed":"gh-changelog-supply","publisher":"github","topics":[],"type":"analysis"},{"id":"7746d02d4673","published":"2025-08-12T16:00:00.002Z","title":"Android’s pKVM Becomes First Globally Certified Software to Achieve Prestigious SESIP Level 5 Security Certification","url":"https://security.googleblog.com/2025/08/Android-pKVM-Certified-SESIP-Level-5.html","summary":"Posted by Dave Kleidermacher, VP Engineering, Android Security & Privacy Today marks a watershed moment and new benchmark for open-source security and the future of consumer electronics. Google is proud to announce that protected KVM (pKVM), the hypervisor that powers the…","feed":"google-security","publisher":"google","topics":[],"type":"analysis"},{"id":"46d17021b485","published":"2025-08-12T16:00:00.000Z","title":"Dow’s 125-year legacy: Innovating with AI to secure a long future","url":"https://microsoft.com/en-us/security/blog/2025/08/12/dows-125-year-legacy-innovating-with-ai-to-secure-a-long-future","summary":"Microsoft recently spoke with Mario Ferket, Chief Information Security Officer for Dow, about the company’s approach to AI in security.","feed":"microsoft-security","publisher":"microsoft","topics":[],"type":"analysis"},{"id":"9512008c137a","published":"2025-08-12T16:00:00.000Z","title":"Android's pKVM hypervisor earns SESIP Level 5 security certification","url":"https://bleepingcomputer.com/news/security/androids-pkvm-hypervisor-earns-sesip-level-5-security-certification","summary":"Google announced that its protected Kernel-based Virtual Machine (pKVM) for Android has achieved SESIP Level 5 certification, the highest security assurance level for IoT and mobile platforms.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["trust"],"type":"analysis"},{"id":"a2ffb45a1d6d","published":"2025-08-12T15:57:22.000Z","title":"Can We Trust AI to Govern AI? Benchmarking LLM Performance on Privacy and AI Governance Exams","url":"https://arxiv.org/abs/2508.09036","summary":"The rapid emergence of large language models (LLMs) has raised urgent questions across the modern workforce about this new technology's strengths, weaknesses, and capabilities. For privacy professionals, the question is whether these AI systems can provide reliable support on…","feed":"arxiv-trust","publisher":"arxiv","topics":["trust"],"type":"research"},{"id":"bef676af13e1","published":"2025-08-12T15:48:53.000Z","title":"NetMoniAI: An Agentic AI Framework for Network Security & Monitoring","url":"https://arxiv.org/abs/2508.10052","summary":"In this paper, we present NetMoniAI, an agentic AI framework for automatic network monitoring and security that integrates decentralized analysis with lightweight centralized coordination. The framework consists of two layers: autonomous micro-agents at each node perform local…","feed":"arxiv-agentic","publisher":"arxiv","topics":[],"type":"research"},{"id":"c76cbe7e5cb9","published":"2025-08-12T15:31:19.000Z","title":"GHSA-222w-xmc5-jhp3 / CVE-2025-43735: Liferay Portal and Liferay DXP have a reflected cross-site scripting vulnerability","url":"https://github.com/advisories/GHSA-222w-xmc5-jhp3","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"53c9b8d0b492","published":"2025-08-12T12:30:32.000Z","title":"GHSA-cg99-m88x-422c / CVE-2025-43736: Liferay Portal and Liferay DXP have a Denial Of Service via File Upload (DOS) vulnerability","url":"https://github.com/advisories/GHSA-cg99-m88x-422c","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"bc211e2c1434","published":"2025-08-12T12:30:32.000Z","title":"GHSA-67mf-3cr5-8w23 / CVE-2025-8885: Bouncy Castle for Java on All (API modules) allows Excessive Allocation","url":"https://github.com/advisories/GHSA-67mf-3cr5-8w23","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"9185d1e9705b","published":"2025-08-12T11:00:00.000Z","title":"The Ultimate Battle: Enterprise Browsers vs. Secure Browser Extensions","url":"https://thehackernews.com/2025/08/the-ultimate-battle-enterprise-browsers.html","summary":"Most security tools can’t see what happens inside the browser, but that’s where the majority of work, and risk, now lives. Security leaders deciding how to close that gap often face a choice: deploy a dedicated Enterprise Browser or add an enterprise-grade control layer to the…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"9670932af165","published":"2025-08-12T09:42:56.000Z","title":"Never compromise with vulnerabilities: a comprehensive survey on AI governance","url":"https://arxiv.org/abs/2508.08789","summary":"The rapid advancement of AI has expanded its capabilities across domains, yet introduced critical technical vulnerabilities, such as algorithmic bias and adversarial sensitivity, that pose significant societal risks, including misinformation, inequity, security breaches,…","feed":"arxiv-trust","publisher":"arxiv","topics":["trust"],"type":"research"},{"id":"dd36286b5d0e","published":"2025-08-12T08:36:00.002Z","title":"Dutch NCSC Confirms Active Exploitation of Citrix NetScaler CVE-2025-6543 in Critical Sectors","url":"https://thehackernews.com/2025/08/dutch-ncsc-confirms-active-exploitation.html","summary":"The Dutch National Cyber Security Centre (NCSC-NL) has warned of cyber attacks exploiting a recently disclosed critical security flaw impacting Citrix NetScaler ADC products to breach organizations in the country. The NCSC-NL said it discovered the exploitation of CVE-2025-6543…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"fd9521a558ff","published":"2025-08-12T06:00:00.000Z","title":"Last Week in AppSec for 12. August 2025","url":"https://checkmarx.com/zero-post/last-week-in-appsec-2025-08-12","summary":"ChatGPT-5 system prompt leaked, CISA supports CVE, and AppSec Village completes; last week in AppSec","feed":"checkmarx","publisher":"checkmarx","topics":[],"type":"analysis"},{"id":"140aaaeedfe4","published":"2025-08-12T03:42:59.000Z","title":"Special-Character Adversarial Attacks on Open-Source Language Model","url":"https://arxiv.org/abs/2508.14070","summary":"Large language models (LLMs) have achieved remarkable performance across diverse natural language processing tasks, yet their vulnerability to character-level adversarial manipulations presents significant security challenges for real-world deployments. This paper presents a…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"8fc7e63fc35a","published":"2025-08-12T00:15:00.000Z","title":"GHSA-r3v7-pc4g-7xp9 / CVE-2025-55152: Oak Server has ReDoS in x-forwarded-proto and x-forwarded-for headers","url":"https://github.com/advisories/GHSA-r3v7-pc4g-7xp9","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"e6fee101e145","published":"2025-08-12T00:14:12.000Z","title":"Securing Agentic AI: Threat Modeling and Risk Analysis for Network Monitoring Agentic AI System","url":"https://arxiv.org/abs/2508.10043","summary":"When combining Large Language Models (LLMs) with autonomous agents, used in network monitoring and decision-making systems, this will create serious security issues. In this research, the MAESTRO framework consisting of the seven layers threat modeling architecture in the…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"e313b968dfcc","published":"2025-08-12T00:13:55.000Z","title":"GHSA-9gvj-pp9x-gcfr / CVE-2025-71325: Picklescan has pickle parsing logic flaw that leads to malicious pickle file bypass","url":"https://github.com/advisories/GHSA-9gvj-pp9x-gcfr","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6eb49eb9db6a","published":"2025-08-12T00:13:46.000Z","title":"GHSA-pwh4-6r3m-j2rf / CVE-2025-55156: PyLoad vulnerable to SQL Injection via API /json/add_package in add_links parameter","url":"https://github.com/advisories/GHSA-pwh4-6r3m-j2rf","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d334d69664c9","published":"2025-08-12T00:13:36.000Z","title":"GHSA-jhmr-57cj-q6g9: Komari vulnerable to 2FA Authentication Bypass","url":"https://github.com/advisories/GHSA-jhmr-57cj-q6g9","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d5043f991390","published":"2025-08-12T00:13:28.000Z","title":"GHSA-q355-h244-969h: Komari vulnerable to Cross-site WebSocket Hijacking","url":"https://github.com/advisories/GHSA-q355-h244-969h","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"3d5ba04b55fc","published":"2025-08-12T00:13:03.000Z","title":"GHSA-xcxh-6cv4-q8p8: HFS user adding a \"web link\" in HFS is vulnerable to \"target=_blank\" exploit","url":"https://github.com/advisories/GHSA-xcxh-6cv4-q8p8","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"278f1521e8fc","published":"2025-08-12T00:00:00.000Z","title":"CVE-2007-0671: Microsoft Office Excel Remote Code Execution Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2007-0671","summary":"Microsoft Office Excel contains a remote code execution vulnerability that can be exploited when a specially crafted Excel file is opened. This malicious file could be delivered as an email attachment or hosted on a malicious website. An attacker could leverage this vulnerability","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"57c3feee641f","published":"2025-08-12T00:00:00.000Z","title":"CVE-2025-8088: RARLAB WinRAR Path Traversal Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2025-8088","summary":"RARLAB WinRAR contains a path traversal vulnerability affecting the Windows version of WinRAR. This vulnerability could allow an attacker to execute arbitrary code by crafting malicious archive files. Known to be used in ransomware campaigns. US federal agencies must act by 2025-","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"ce815bbbf68c","published":"2025-08-12T00:00:00.000Z","title":"CVE-2013-3893: Microsoft Internet Explorer Resource Management Errors Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2013-3893","summary":"Microsoft Internet Explorer contains a memory corruption vulnerability that allows for remote code execution. The impacted products could be end-of-life (EoL) and/or end-of-service (EoS). Users should discontinue product utilization. US federal agencies must act by 2025-09-02.","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"f990363ec8c8","published":"2025-08-12T00:00:00.000Z","title":"Agentic Frameworks Summary","url":"https://elastic.co/security-labs/blog/agentic-ai-summary","summary":"Agentic systems require security teams to balance autonomy with alignment, ensuring that AI agents can act independently while remaining goal-consistent and controllable .","feed":"elastic-labs","publisher":"elastic","topics":["trust","ai-security"],"type":"analysis"},{"id":"652495f72723","published":"2025-08-11T23:07:36.000Z","title":"GHSA-674p-xv2x-rf3g: Litestar has potential log injection in exception logging","url":"https://github.com/advisories/GHSA-674p-xv2x-rf3g","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ce25fb8443f3","published":"2025-08-11T23:02:11.000Z","title":"Model Context Protocol SEP-1330: Elicitation Enum Schema Improvements and Standards Compliance","url":"https://github.com/modelcontextprotocol/modelcontextprotocol/issues/1330","summary":"Final specification change. Preamble Title: Elicitation Enum Schema Improvements and Standards Compliance Author(s): Cliff Hall (@cliffhall), Tapan Chugh (@chughtapan) Track: Standards Track Status: Proposal Created: 2025-08-11 Abstract This SEP proposes improvements to enum…","feed":"mcp-seps","publisher":"mcp","topics":["trust"],"type":"proposal"},{"id":"159e7468ced6","published":"2025-08-11T22:45:20.000Z","title":"GHSA-qx2v-8332-m4fv / CVE-2025-55159: slab allows out-of-bounds access in `get_disjoint_mut` due to incorrect bounds check","url":"https://github.com/advisories/GHSA-qx2v-8332-m4fv","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"96118a8382ad","published":"2025-08-11T22:12:27.000Z","title":"FIDELIS: Blockchain-Enabled Protection Against Poisoning Attacks in Federated Learning","url":"https://arxiv.org/abs/2508.10042","summary":"Federated learning enhances traditional deep learning by enabling the joint training of a model with the use of IoT device's private data. It ensures privacy for clients, but is susceptible to data poisoning attacks during training that degrade model performance and integrity.…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"18d45bfc3b58","published":"2025-08-11T21:31:40.000Z","title":"GHSA-v6c8-g53h-mc2h / CVE-2025-53910: Mattermost Confluence Plugin has Missing Authorization vulnerability","url":"https://github.com/advisories/GHSA-v6c8-g53h-mc2h","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"220132d647f4","published":"2025-08-11T21:31:40.000Z","title":"GHSA-3cg3-3mmr-w8hj / CVE-2025-54525: Mattermost Confluence Plugin has Improper Validation of Specified Type of Input","url":"https://github.com/advisories/GHSA-3cg3-3mmr-w8hj","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"37e35dae9974","published":"2025-08-11T21:31:40.000Z","title":"GHSA-qjrx-j8wm-xf83 / CVE-2025-8285: Mattermost Confluence Plugin has Missing Authorization vulnerability","url":"https://github.com/advisories/GHSA-qjrx-j8wm-xf83","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"533547ce9b94","published":"2025-08-11T21:31:40.000Z","title":"GHSA-qpjq-c5hr-7925 / CVE-2025-54478: Mattermost Confluence Plugin is Missing Authentication for Critical Function","url":"https://github.com/advisories/GHSA-qpjq-c5hr-7925","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"7bedff4839f7","published":"2025-08-11T21:31:40.000Z","title":"GHSA-w92j-c6gr-hj8r / CVE-2025-53514: Mattermost Confluence Plugin has Improper Check for Unusual or Exceptional Conditions","url":"https://github.com/advisories/GHSA-w92j-c6gr-hj8r","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"7d8809e58a1d","published":"2025-08-11T21:31:40.000Z","title":"GHSA-vc77-c2hx-h5x2 / CVE-2025-52931: Mattermost Confluence Plugin has Improper Check for Unusual or Exceptional Conditions","url":"https://github.com/advisories/GHSA-vc77-c2hx-h5x2","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"8d7e7f5cdbaa","published":"2025-08-11T21:31:40.000Z","title":"GHSA-gjpm-6w34-ppvf / CVE-2025-54463: Mattermost Confluence Plugin has Improper Check for Unusual or Exceptional Conditions","url":"https://github.com/advisories/GHSA-gjpm-6w34-ppvf","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b6cc9d165cac","published":"2025-08-11T21:31:40.000Z","title":"GHSA-42m6-5vm7-fjv2 / CVE-2025-53857: Mattermost Confluence Plugin has Missing Authorization vulnerability","url":"https://github.com/advisories/GHSA-42m6-5vm7-fjv2","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"fbeb43ea122a","published":"2025-08-11T21:31:40.000Z","title":"GHSA-j66h-xhpr-7q5g / CVE-2025-54458: Mattermost Confluence Plugin has Missing Authorization vulnerability","url":"https://github.com/advisories/GHSA-j66h-xhpr-7q5g","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"045e3fab9e28","published":"2025-08-11T21:31:39.000Z","title":"GHSA-cmpr-8prq-w5p5 / CVE-2025-48731: Mattermost Confluence Plugin has Missing Authorization vulnerability","url":"https://github.com/advisories/GHSA-cmpr-8prq-w5p5","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6b1c9e1e9c6e","published":"2025-08-11T21:31:39.000Z","title":"GHSA-rfg4-2m63-fw2q / CVE-2025-49221: Mattermost Confluence Plugin has Missing Authorization vulnerability","url":"https://github.com/advisories/GHSA-rfg4-2m63-fw2q","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"b5755c246da6","published":"2025-08-11T21:31:39.000Z","title":"GHSA-6ff3-jgxh-vffj / CVE-2025-44004: Mattermost Confluence Plugin is Missing Authentication for Critical Function","url":"https://github.com/advisories/GHSA-6ff3-jgxh-vffj","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"e799204c38fb","published":"2025-08-11T21:31:38.000Z","title":"GHSA-vpcr-fqpc-386h / CVE-2025-44001: Mattermost Confluence Plugin has Missing Authorization vulnerability","url":"https://github.com/advisories/GHSA-vpcr-fqpc-386h","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"bddc53364abc","published":"2025-08-11T21:27:58.000Z","title":"SonicWall pins firewall attack spree on year-old vulnerability","url":"https://cyberscoop.com/sonicwall-firewall-attacks-old-vulnerability","summary":"The vendor ruled out a zero-day vulnerability as the root cause, disputing initial assessments from third-party researchers. Fewer than 40 organizations have been impacted since mid-July.","feed":"cyberscoop","publisher":"cyberscoop","topics":[],"type":"research"},{"id":"5ed7a970d985","published":"2025-08-11T19:19:00.000Z","title":"Netherlands: Citrix Netscaler flaw CVE-2025-6543 exploited to breach orgs","url":"https://bleepingcomputer.com/news/security/netherlands-citrix-netscaler-flaw-cve-2025-6543-exploited-to-breach-orgs","summary":"The Netherlands' National Cyber Security Centre (NCSC) is warning that a critical Citrix NetScaler vulnerability tracked as CVE-2025-6543 was exploited to breach \"critical organizations\" in the country.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"6256ff5cdf88","published":"2025-08-11T18:05:00.000Z","title":"Details emerge on WinRAR zero-day attacks that infected PCs with malware","url":"https://bleepingcomputer.com/news/security/details-emerge-on-winrar-zero-day-attacks-that-infected-pcs-with-malware","summary":"Researchers have released a report detailing how a recent WinRAR path traversal vulnerability tracked as CVE-2025-8088 was exploited in zero-day attacks by the Russian 'RomCom' hacking group to drop different malware payloads.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"research"},{"id":"f9fb61d642d7","published":"2025-08-11T16:50:00.000Z","title":"Microsoft tests cloud-based Windows 365 disaster recovery PCs","url":"https://bleepingcomputer.com/news/microsoft/microsoft-tests-cloud-based-windows-365-disaster-recovery-pcs","summary":"Microsoft has announced the limited public preview of Windows 365 Reserve, a service that provides temporary desktop access to pre-configured cloud PCs for employees whose computers have become unavailable due to cyberattacks, hardware issues, or software problems.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"12d29417552d","published":"2025-08-11T16:04:47.000Z","title":"BlindGuard: Safeguarding LLM-based Multi-Agent Systems under Unknown Attacks","url":"https://arxiv.org/abs/2508.08127","summary":"The security of LLM-based multi-agent systems (MAS) is critically threatened by propagation vulnerability, where malicious agents can distort collective decision-making through inter-agent message interactions. While existing supervised defense methods demonstrate promising…","feed":"arxiv-aisec","publisher":"arxiv","topics":[],"type":"research"},{"id":"a8cda4ff47a7","published":"2025-08-11T15:17:00.000Z","title":"The Rise of Native Phishing: Microsoft 365 Apps Abused in Attacks","url":"https://bleepingcomputer.com/news/security/the-rise-of-native-phishing-microsoft-365-apps-abused-in-attacks","summary":"Native phishing turns trusted tools into attack delivery systems. Varonis shows how attackers weaponize Microsoft 365 apps, like OneNote & OneDrive, to send convincing internal lures and how to spot them before they spread.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"01a437c474d4","published":"2025-08-11T15:08:00.002Z","title":"Researchers Spot Surge in Erlang/OTP SSH RCE Exploits, 70% Target OT Firewalls","url":"https://thehackernews.com/2025/08/researchers-spot-surge-in-erlangotp-ssh.html","summary":"Malicious actors have been observed exploiting a now-patched critical security flaw impacting Erlang/Open Telecom Platform (OTP) SSH as early as beginning of May 2025, with about 70% of detections originating from firewalls protecting operational technology (OT) networks. The…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"research"},{"id":"ff85433135ee","published":"2025-08-11T14:42:44.000Z","title":"BadPromptFL: A Novel Backdoor Threat to Prompt-based Federated Learning in Multimodal Models","url":"https://arxiv.org/abs/2508.08040","summary":"Prompt-based tuning has emerged as a lightweight alternative to full fine-tuning in large vision-language models, enabling efficient adaptation via learned contextual prompts. This paradigm has recently been extended to federated learning settings (e.g., PromptFL), where…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"aefee43254ca","published":"2025-08-11T13:38:14.000Z","title":"GHSA-rrgf-hcr9-jq6h / CVE-2025-55149: TinyScientist has Path Traversal Vulnerability in PDF Review Function (CWE-22)","url":"https://github.com/advisories/GHSA-rrgf-hcr9-jq6h","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"8b4001f9eff7","published":"2025-08-11T11:53:00.001Z","title":"⚡ Weekly Recap: BadCam Attack, WinRAR 0-Day, EDR Killer, NVIDIA Flaws, Ransomware Attacks & More","url":"https://thehackernews.com/2025/08/weekly-recap-badcam-attack-winrar-0-day.html","summary":"This week, cyber attackers are moving quickly, and businesses need to stay alert. They’re finding new weaknesses in popular software and coming up with clever ways to get around security. Even one unpatched flaw could let attackers in, leading to data theft or even taking…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"roundup"},{"id":"cca582ac9386","published":"2025-08-11T11:20:58.000Z","title":"Claude Code: Data Exfiltration with DNS (CVE-2025-55284)","url":"https://embracethered.com/blog/posts/2025/claude-code-exfiltration-via-dns-requests","summary":"Today we cover Claude Code and a high severity vulnerability that Anthropic fixed in early June. The vulnerability allowed an attacker to hijack Claude Code via indirect prompt injection and leak sensitive information from the developer’s machine, e.g. API keys, to external…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","remediation","ai-security"],"type":"analysis"},{"id":"c209aa7fedde","published":"2025-08-11T10:00:48.000Z","title":"Keys to the Kingdom: Erlang/OTP SSH Vulnerability Analysis and Exploits Observed in the Wild","url":"https://unit42.paloaltonetworks.com/erlang-otp-cve-2025-32433","summary":"CVE-2025-32433 allows for remote code execution in sshd for certain versions of Erlang programming language’s OTP. We reproduced this CVE and share our findings.","feed":"unit42","publisher":"palo-alto","topics":[],"type":"analysis"},{"id":"98fccff20ef0","published":"2025-08-11T09:35:00.000Z","title":"Over 29,000 Exchange servers unpatched against high-severity flaw","url":"https://bleepingcomputer.com/news/security/over-29-000-exchange-servers-unpatched-against-high-severity-flaw","summary":"Over 29,000 Exchange servers exposed online remain unpatched against a high-severity vulnerability that can let attackers move laterally in Microsoft cloud environments, potentially leading to complete domain compromise.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"3c5cbc7d56b3","published":"2025-08-11T09:23:51.000Z","title":"IDC Commentary on AI-Assisted Development is a Wake-Up Call for AppSec Leaders","url":"https://checkmarx.com/blog/ai-llm-tools-in-application-security/idc-commentary-on-ai-assisted-development-is-a-wake-up-call-for-appsec-leaders","summary":"The latest IDC Link commentary on Checkmarx’s Agentic AI solution and the security risks of Vibe Coding shines a bright light on the fact that pre-AI application security is no longer sustainable in the era of AI-generated code and code vibing. In the IDC Link, ‘Checkmarx…","feed":"checkmarx","publisher":"checkmarx","topics":[],"type":"analysis"},{"id":"6cb7a8d756a0","published":"2025-08-11T05:57:41.000Z","title":"Multi-Turn Jailbreaks Are Simpler Than They Seem","url":"https://arxiv.org/abs/2508.07646","summary":"While defenses against single-turn jailbreak attacks on Large Language Models (LLMs) have improved significantly, multi-turn jailbreaks remain a persistent vulnerability, often achieving success rates exceeding 70% against models optimized for single-turn protection. This work…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"061f362817a9","published":"2025-08-11T05:54:00.008Z","title":"WinRAR Zero-Day Under Active Exploitation – Update to Latest Version Immediately","url":"https://thehackernews.com/2025/08/winrar-zero-day-under-active.html","summary":"The maintainers of the WinRAR file archiving utility have released an update to address an actively exploited zero-day vulnerability. Tracked as CVE-2025-8088 (CVSS score: 8.8), the issue has been described as a case of path traversal affecting the Windows version of the tool…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"fdbc4e25aa4a","published":"2025-08-10T12:31:00.002Z","title":"Researchers Detail Windows EPM Poisoning Exploit Chain Leading to Domain Privilege Escalation","url":"https://thehackernews.com/2025/08/researchers-detail-windows-epm.html","summary":"Cybersecurity researchers have presented new findings related to a now-patched security issue in Microsoft's Windows Remote Procedure Call (RPC) communication protocol that could be abused by an attacker to conduct spoofing attacks and impersonate a known server. The…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"research"},{"id":"f22762e48c01","published":"2025-08-10T11:20:58.000Z","title":"ZombAI Exploit with OpenHands: Prompt Injection To Remote Code Execution","url":"https://embracethered.com/blog/posts/2025/openhands-remote-code-execution-zombai","summary":"Today we have another post about OpenHands from All Hands AI. It is a popular agent, initially named “OpenDevin”, and recently the company also provides a cloud-based service . Which is all pretty cool and exciting. Prompt Injection to Full System Compromise However, as you…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["ai-security"],"type":"analysis"},{"id":"548df3046dec","published":"2025-08-10T06:39:11.000Z","title":"WhyFlow: Interrogative Debugger for Sensemaking Taint Analysis","url":"https://arxiv.org/abs/2508.07198","summary":"Taint analysis is a security analysis technique used to track the flow of potentially dangerous data through an application and its dependent libraries. Investigating why certain unexpected flows appear and why expected flows are missing is an important sensemaking process…","feed":"arxiv-sca","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"3278564729fb","published":"2025-08-10T01:59:07.000Z","title":"A Real-Time, Self-Tuning Moderator Framework for Adversarial Prompt Detection","url":"https://arxiv.org/abs/2508.07139","summary":"Ensuring LLM alignment is critical to information security as AI models become increasingly widespread and integrated in society. Unfortunately, many defenses against adversarial attacks and jailbreaking on LLMs cannot adapt quickly to new attacks, degrade model responses to…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"4f0442fae7fd","published":"2025-08-09T23:30:03.000Z","title":"Prompt injection engineering for attackers: Exploiting GitHub Copilot","url":"https://blog.trailofbits.com/2025/08/06/prompt-injection-engineering-for-attackers-exploiting-github-copilot","summary":"Hacker News: 11 points, 1 comments.","feed":"hn-ai-security-0","publisher":"hacker-news","topics":["agentic-sdlc","ai-security"],"type":"analysis"},{"id":"9fec88be087f","published":"2025-08-09T20:38:07.000Z","title":"ScamDetect: Towards a Robust, Agnostic Framework to Uncover Threats in Smart Contracts","url":"https://arxiv.org/abs/2508.07094","summary":"Smart contracts have transformed decentralized finance by enabling programmable, trustless transactions. However, their widespread adoption and growing financial significance have attracted persistent and sophisticated threats, such as phishing campaigns and contract-level…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"91d0df3c31b0","published":"2025-08-09T18:55:00.000Z","title":"Researchers Reveal ReVault Attack Targeting Dell ControlVault3 Firmware in 100+ Laptop Models","url":"https://thehackernews.com/2025/08/researchers-reveal-revault-attack.html","summary":"Cybersecurity researchers have uncovered multiple security flaws in Dell's ControlVault3 firmware and its associated Windows APIs that could have been abused by attackers to bypass Windows login, extract cryptographic keys, as well as maintain access even after a fresh…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"research"},{"id":"8a22563d6e71","published":"2025-08-09T15:06:00.003Z","title":"Researchers Uncover GPT-5 Jailbreak and Zero-Click AI Agent Attacks Exposing Cloud and IoT Systems","url":"https://thehackernews.com/2025/08/researchers-uncover-gpt-5-jailbreak-and.html","summary":"Cybersecurity researchers have uncovered a jailbreak technique to bypass ethical guardrails erected by OpenAI in its latest large language model (LLM) GPT-5 and produce illicit instructions. Generative artificial intelligence (AI) security platform NeuralTrust said it combined…","feed":"thehackernews","publisher":"thehackernews","topics":["models","ai-security"],"type":"research"},{"id":"c10319285b39","published":"2025-08-09T14:45:53.000Z","title":"Context Engineering for Multi-Agent LLM Code Assistants Using Elicit, NotebookLM, ChatGPT, and Claude Code","url":"https://arxiv.org/abs/2508.08322","summary":"Large Language Models (LLMs) have shown promise in automating code generation and software engineering tasks, yet they often struggle with complex, multi-file projects due to context limitations and knowledge gaps. We propose a novel context engineering workflow that combines…","feed":"arxiv-agentic","publisher":"arxiv","topics":["agentic-sdlc"],"type":"research"},{"id":"e6a7fd0309c3","published":"2025-08-09T14:11:00.000Z","title":"60 malicious Ruby gems downloaded 275,000 times steal credentials","url":"https://bleepingcomputer.com/news/security/60-malicious-ruby-gems-downloaded-275-000-times-steal-credentials","summary":"Sixty malicious Ruby gems containing credential-stealing code have been downloaded over 275,000 times since March 2023, targeting developer accounts.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["sca"],"type":"analysis"},{"id":"a02d31cafa20","published":"2025-08-09T10:00:58.000Z","title":"OpenHands and the Lethal Trifecta: How Prompt Injection Can Leak Access Tokens","url":"https://embracethered.com/blog/posts/2025/openhands-the-lethal-trifecta-strikes-again","summary":"Another day, another AI data exfiltration exploit. Today we talk about OpenHands , formerly referred to as OpenDevin. It’s created by All-Hands AI. The OpenHands agent renders images in chat, which enables zero-click data exfiltration. Simon Willison recently gave this data…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["ai-security"],"type":"incident"},{"id":"25f21f8c5c91","published":"2025-08-09T06:30:28.000Z","title":"GHSA-c6g5-g6r7-q4j6 / CVE-2025-4655: Liferay Portal and Liferay DXP vulnerable to Server-Side Request Forgery","url":"https://github.com/advisories/GHSA-c6g5-g6r7-q4j6","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"64966b2d7a52","published":"2025-08-09T06:30:28.000Z","title":"GHSA-6v93-frf9-2rp8 / CVE-2025-4581: Liferay Portal and Liferay DXP vulnerable to Server-Side Request Forgery","url":"https://github.com/advisories/GHSA-6v93-frf9-2rp8","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d2910980cc4c","published":"2025-08-09T05:19:04.000Z","title":"When a Jira Ticket Can Steal Your Secrets","url":"https://simonwillison.net/2025/Aug/9/when-a-jira-ticket-can-steal-your-secrets","summary":"When a Jira Ticket Can Steal Your Secrets Zenity Labs describe a classic lethal trifecta attack, this time against Cursor, MCP, Jira and Zendesk. They also have a short video demonstrating the issue . Zendesk support emails are often connected to Jira, such that incoming…","feed":"simonw","publisher":"simonw","topics":["ai-security"],"type":"analysis"},{"id":"0d308becda5e","published":"2025-08-09T04:30:36.000Z","title":"My Lethal Trifecta talk at the Bay Area AI Security Meetup","url":"https://simonwillison.net/2025/Aug/9/bay-area-ai","summary":"I gave a talk on Wednesday at the Bay Area AI Security Meetup about prompt injection, the lethal trifecta and the challenges of securing systems that use MCP. It wasn't recorded but I've created an annotated presentation with my slides and detailed notes on everything I talked…","feed":"simonw","publisher":"simonw","topics":["ai-security"],"type":"analysis"},{"id":"54dc2271b4af","published":"2025-08-09T02:37:59.000Z","title":"Context Misleads LLMs: The Role of Context Filtering in Maintaining Safe Alignment of LLMs","url":"https://arxiv.org/abs/2508.10031","summary":"While Large Language Models (LLMs) have shown significant advancements in performance, various jailbreak attacks have posed growing safety and ethical risks. Malicious users often exploit adversarial context to deceive LLMs, prompting them to generate responses to harmful…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"0dc64308a4e1","published":"2025-08-09T00:10:42.000Z","title":"Nuclei Templates releases Nuclei Templates v10.2.7 - Release Notes","url":"https://github.com/projectdiscovery/nuclei-templates/releases/tag/v10.2.7","summary":"New Templates Added: 50 | CVEs Added: 08 | First-time contributions: 3 🔥 Release Highlights 🔥 [CVE-2025-54782] NestJS DevTools Integration - Remote Code Execution (@nukunga) [critical] 🔥 [CVE-2025-25257] Fortinet FortiWeb - SQL Injection (@watchtowr, @johnk3r) [critical] 🔥…","feed":"nuclei-templates","publisher":"projectdiscovery","topics":["dast"],"type":"release"},{"id":"57998aed0323","published":"2025-08-08T23:22:02.000Z","title":"Friday Squid Blogging: New Vulnerability in Squid HTTP Proxy Server","url":"https://schneier.com/blog/archives/2025/08/friday-squid-blogging-new-vulnerability-in-squid-http-proxy-server.html","summary":"In a rare squid/security combined post, a new vulnerability was discovered in the Squid HTTP proxy server. Read More →","feed":"schneier","publisher":"schneier","topics":[],"type":"analysis"},{"id":"17230540e546","published":"2025-08-08T21:53:16.000Z","title":"DARPA’s AI Cyber Challenge reveals winning models for automated vulnerability discovery and patching","url":"https://cyberscoop.com/darpa-ai-cyber-challenge-winners-def-con-2025","summary":"The initiative seeks to patch vulnerabilities in open-source code before they are exploited by would-be attackers. Now comes the hard part — putting the systems to the test in the real world.","feed":"cyberscoop","publisher":"cyberscoop","topics":["remediation"],"type":"analysis"},{"id":"184f42a6a17f","published":"2025-08-08T20:42:00.000Z","title":"WinRAR zero-day exploited to plant malware on archive extraction","url":"https://bleepingcomputer.com/news/security/winrar-zero-day-flaw-exploited-by-romcom-hackers-in-phishing-attacks","summary":"A recently fixed WinRAR vulnerability tracked as CVE-2025-8088 was exploited as a zero-day in phishing attacks to install the RomCom malware.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"32e8f215265f","published":"2025-08-08T19:32:50.000Z","title":"GHSA-2vcf-qxv3-2mgw / CVE-2025-54417: Craft CMS has a theoretical bypass for CVE-2025-23209","url":"https://github.com/advisories/GHSA-2vcf-qxv3-2mgw","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"7fda8fbcec4c","published":"2025-08-08T19:03:45.000Z","title":"From Beginner to Builder: Understanding OpenSSF Community and Working Groups","url":"https://openssf.org/blog/2025/08/08/from-beginner-to-builder-understanding-openssf-community-and-working-groups","summary":"The Open Source Security Foundation (OpenSSF) serves as the global hub for collaborative work on securing the software supply chain. Whether you’re an open-source maintainer, a security engineer, a student, or someone passionate about public digital infrastructure, OpenSSF…","feed":"openssf","publisher":"openssf","topics":["sca"],"type":"analysis"},{"id":"b2017114b485","published":"2025-08-08T18:32:22.000Z","title":"GHSA-6qcg-28jh-hm7r / CVE-2025-4576: Liferay Portal Reflected XSS in blogs-web","url":"https://github.com/advisories/GHSA-6qcg-28jh-hm7r","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"5a9cda219055","published":"2025-08-08T17:29:16.000Z","title":"Latent Fusion Jailbreak: Blending Harmful and Harmless Representations to Elicit Unsafe LLM Outputs","url":"https://arxiv.org/abs/2508.10029","summary":"Safety-aligned large language models can still be manipulated through white-box interventions that modify their internal representations. We introduce Latent Fusion Jailbreak (LFJ), which works by pairing a harmful query with a structurally similar but benign counterpart, then…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"f5ea4707db29","published":"2025-08-08T17:08:46.000Z","title":"GHSA-v3gr-w9gf-23cx / CVE-2025-55009: The AuthKit Remix Library renders sensitive auth data in HTML","url":"https://github.com/advisories/GHSA-v3gr-w9gf-23cx","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"f9e43beb4b06","published":"2025-08-08T17:04:11.000Z","title":"GHSA-vqvc-9q8x-vmq6 / CVE-2025-55008: The AuthKit React Router Library rendered sensitive auth data in HTML","url":"https://github.com/advisories/GHSA-vqvc-9q8x-vmq6","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"351422c61e0e","published":"2025-08-08T17:01:41.000Z","title":"ScamAgents: How AI Agents Can Simulate Human-Level Scam Calls","url":"https://arxiv.org/abs/2508.06457","summary":"Large Language Models (LLMs) have demonstrated impressive fluency and reasoning capabilities, but their potential for misuse has raised growing concern. In this paper, we present ScamAgent, an autonomous multi-turn agent built on top of LLMs, capable of generating highly…","feed":"arxiv-agentic","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"bb5a36cd34c0","published":"2025-08-08T16:14:00.002Z","title":"AI Tools Fuel Brazilian Phishing Scam While Efimer Trojan Steals Crypto from 5,000 Victims","url":"https://thehackernews.com/2025/08/ai-tools-fuel-brazilian-phishing-scam.html","summary":"Cybersecurity researchers are drawing attention to a new campaign that's using legitimate generative artificial intelligence (AI)-powered website building tools like DeepSite AI and BlackBox AI to create replica phishing pages mimicking Brazilian government agencies as part of…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"research"},{"id":"4159bd3c8fa9","published":"2025-08-08T16:13:28.000Z","title":"Learning to Detect Unseen Jailbreak Attacks in Large Vision-Language Models","url":"https://arxiv.org/abs/2508.09201","summary":"Despite extensive alignment efforts, Large Vision-Language Models (LVLMs) remain vulnerable to jailbreak attacks. To mitigate these risks, existing detection methods are essential, yet they face two major challenges: generalization and accuracy. While learning-based methods…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"c8f94bdb8bef","published":"2025-08-08T15:17:09.000Z","title":"GHSA-2q8q-8fgw-9p6p / CVE-2025-55001: OpenBao LDAP MFA Enforcement Bypass When Using Username As Alias","url":"https://github.com/advisories/GHSA-2q8q-8fgw-9p6p","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6fbdee6ccb33","published":"2025-08-08T14:44:49.000Z","title":"GHSA-rxp7-9q75-vj3p / CVE-2025-55003: OpenBao Login MFA Bypass of Rate Limiting and TOTP Token Reuse","url":"https://github.com/advisories/GHSA-rxp7-9q75-vj3p","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6df2e49577e3","published":"2025-08-08T14:42:49.000Z","title":"GHSA-f7c3-mhj2-9pvg / CVE-2025-55000: OpenBao TOTP Secrets Engine Code Reuse","url":"https://github.com/advisories/GHSA-f7c3-mhj2-9pvg","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ff2b82ef9de7","published":"2025-08-08T14:41:14.000Z","title":"GHSA-hh28-h22f-8357 / CVE-2025-54999: OpenBao has a Timing Side-Channel in the Userpass Auth Method","url":"https://github.com/advisories/GHSA-hh28-h22f-8357","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d6d62572e1fc","published":"2025-08-08T14:38:54.000Z","title":"GHSA-j3xv-7fxp-gfhx / CVE-2025-54998: OpenBao Userpass and LDAP User Lockout Bypass","url":"https://github.com/advisories/GHSA-j3xv-7fxp-gfhx","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"50c2fb304bcc","published":"2025-08-08T14:37:22.000Z","title":"GHSA-xp75-r577-cvhp / CVE-2025-54997: Privileged OpenBao Operator May Execute Code on the Underlying Host","url":"https://github.com/advisories/GHSA-xp75-r577-cvhp","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"9b95de3dec49","published":"2025-08-08T14:32:42.000Z","title":"GHSA-vf84-mxrq-crqc / CVE-2025-54996: OpenBao Root Namespace Operator May Elevate Token Privileges","url":"https://github.com/advisories/GHSA-vf84-mxrq-crqc","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"395c1b05af40","published":"2025-08-08T14:29:48.000Z","title":"GHSA-6jcc-xgcr-q3h4 / CVE-2025-54888: @fedify/fedify has Improper Authentication and Incorrect Authorization","url":"https://github.com/advisories/GHSA-6jcc-xgcr-q3h4","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"e7a3c6006d91","published":"2025-08-08T13:15:40.000Z","title":"LLM Robustness Leaderboard v1 --Technical report","url":"https://arxiv.org/abs/2508.06296","summary":"This technical report accompanies the LLM robustness leaderboard published by PRISM Eval for the Paris AI Action Summit. We introduce PRISM Eval Behavior Elicitation Tool (BET), an AI system performing automated red-teaming through Dynamic Adversarial Optimization that achieves…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"05f0e405aa72","published":"2025-08-08T12:32:17.000Z","title":"GHSA-g4px-6qhm-hqjm / CVE-2025-48913: Apache CXF: Untrusted JMS configuration can lead to RCE","url":"https://github.com/advisories/GHSA-g4px-6qhm-hqjm","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"4a4e56e6fbac","published":"2025-08-08T12:32:17.000Z","title":"GHSA-g358-g2pq-c46j / CVE-2025-53606: Apache Seata: Deserialization of untrusted Data in Apache Seata Server","url":"https://github.com/advisories/GHSA-g358-g2pq-c46j","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"448561e0ee82","published":"2025-08-08T11:01:14.000Z","title":"Google Project Zero Changes Its Disclosure Policy","url":"https://schneier.com/blog/archives/2025/08/google-project-zero-changes-its-disclosure-policy.html","summary":"Google’s vulnerability finding team is again pushing the envelope of responsible disclosure: Google’s Project Zero team will retain its existing 90+30 policy regarding vulnerability disclosures, in which it provides vendors with 90 days before full disclosure takes place, with…","feed":"schneier","publisher":"schneier","topics":["remediation"],"type":"standard"},{"id":"f89c4b67c6ca","published":"2025-08-08T10:58:00.004Z","title":"RubyGems, PyPI Hit by Malicious Packages Stealing Credentials, Crypto, Forcing Security Changes","url":"https://thehackernews.com/2025/08/rubygems-pypi-hit-by-malicious-packages.html","summary":"A fresh set of 60 malicious packages has been uncovered targeting the RubyGems ecosystem by posing as seemingly innocuous automation tools for social media, blogging, or messaging services to steal credentials from unsuspecting users and likely resell them on dark web forums…","feed":"thehackernews","publisher":"thehackernews","topics":["sca"],"type":"research"},{"id":"f6df5da2ae46","published":"2025-08-08T09:17:33.000Z","title":"SLIP: Soft Label Mechanism and Key-Extraction-Guided CoT-based Defense Against Instruction Backdoor in APIs","url":"https://arxiv.org/abs/2508.06153","summary":"Customized Large Language Model (LLM) agents face a critical security threat from black-box instruction backdoors, where malicious behaviors are covertly injected through hidden system instructions. Although existing prompt-based defenses can often detect poisoned inputs, they…","feed":"arxiv-agentic","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"8824dde8a7ae","published":"2025-08-08T07:16:00.002Z","title":"GreedyBear Steals $1M in Crypto Using 150+ Malicious Firefox Wallet Extensions","url":"https://thehackernews.com/2025/08/greedybear-steals-1m-in-crypto-using.html","summary":"A newly discovered campaign dubbed GreedyBear has leveraged over 150 malicious extensions to the Firefox marketplace that are designed to impersonate popular cryptocurrency wallets and steal more than $1 million in digital assets. The published browser add-ons masquerade as…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"ad52a5a9ec47","published":"2025-08-08T07:02:58.000Z","title":"AI Kill Chain in Action: Devin AI Exposes Ports to the Internet with Prompt Injection","url":"https://embracethered.com/blog/posts/2025/devin-ai-kill-chain-exposing-ports","summary":"Today let’s explore Devin’s system prompt a bit more. Specifically, an interesing tool that I discovered when reading through it. Hidden in Devin’s capabilities is a tool that can open any local port to the public Internet. That means, with the right indirect prompt injection…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["ai-security"],"type":"research"},{"id":"009da03ee0a2","published":"2025-08-08T03:20:25.000Z","title":"Multi-Level Safety Continual Projection for Fine-Tuned Large Language Models without Retraining","url":"https://arxiv.org/abs/2508.09190","summary":"While fine-tuning services drive the rapid expansion of task capabilities in large language models (LLMs), they are often accompanied by the degradation and reorganization of safety-aligned representations, making models more prone to deviating from human preferences and…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"5490a754819c","published":"2025-08-08T00:30:26.000Z","title":"GHSA-xc7w-r669-48pf / CVE-2025-54951: ExecuTorch vulnerable to Heap-based Buffer Overflow","url":"https://github.com/advisories/GHSA-xc7w-r669-48pf","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"58f44c9f1182","published":"2025-08-08T00:30:26.000Z","title":"GHSA-9m39-3mf3-xwch / CVE-2025-54949: ExecuTorch heap buffer overflow vulnerability","url":"https://github.com/advisories/GHSA-9m39-3mf3-xwch","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"8116f8a1f7bc","published":"2025-08-08T00:30:26.000Z","title":"GHSA-84m3-f99p-cqx5 / CVE-2025-30405: ExecuTorch integer overflow vulnerability","url":"https://github.com/advisories/GHSA-84m3-f99p-cqx5","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a6fccad7d07b","published":"2025-08-08T00:30:26.000Z","title":"GHSA-33r8-vrx9-rmcv / CVE-2025-54952: ExecuTorch integer overflow vulnerability leads to code execution","url":"https://github.com/advisories/GHSA-33r8-vrx9-rmcv","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c33d8b77aabb","published":"2025-08-08T00:30:26.000Z","title":"GHSA-f9hx-c6jf-3qxm / CVE-2025-54950: ExecuTorch out-of-bounds access vulnerability","url":"https://github.com/advisories/GHSA-f9hx-c6jf-3qxm","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"da2e96f8f181","published":"2025-08-08T00:30:26.000Z","title":"GHSA-hj95-mhgf-jxc4 / CVE-2025-30404: ExecuTorch integer overflow vulnerability","url":"https://github.com/advisories/GHSA-hj95-mhgf-jxc4","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"f9bbff5f9860","published":"2025-08-08T00:00:00.000Z","title":"Building a Secure Software Supply Chain: 5 Key Insights from GitLab’s Field CTO","url":"https://chainguard.dev/unchained/building-a-secure-software-supply-chain-5-key-insights-from-gitlabs-field-cto","summary":"Chainguard and GitLab recently recorded a webinar discussing challenges organizations face in building a secure software supply chain. Get the key takeaways.","feed":"chainguard","publisher":"chainguard","topics":["sca"],"type":"analysis"},{"id":"2e7a01178e89","published":"2025-08-07T21:42:00.000Z","title":"Fake WhatsApp developer libraries hide destructive data-wiping code","url":"https://bleepingcomputer.com/news/security/fake-whatsapp-developer-libraries-hide-destructive-data-wiping-code","summary":"Two malicious NPM packages posing as WhatsApp development tools have been discovered deploying destructive data-wiping code that recursively deletes files on a developer's computers.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["sca"],"type":"analysis"},{"id":"0fadbe26c602","published":"2025-08-07T21:31:08.000Z","title":"GHSA-856v-8qm2-9wjv / CVE-2025-7195: operator-sdk: privilege escalation due to incorrect permissions of /etc/passwd","url":"https://github.com/advisories/GHSA-856v-8qm2-9wjv","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"be470dc6a2b5","published":"2025-08-07T20:55:35.000Z","title":"GHSA-c7p4-hx26-pr73 / CVE-2025-54887: JWE is missing AES-GCM authentication tag validation in encrypted JWE","url":"https://github.com/advisories/GHSA-c7p4-hx26-pr73","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"69db55bc9574","published":"2025-08-07T20:53:40.000Z","title":"GHSA-m3hh-f9gh-74c2 / CVE-2025-7054: quiche connection ID retirement can trigger an infinite loop","url":"https://github.com/advisories/GHSA-m3hh-f9gh-74c2","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"66b112920bdf","published":"2025-08-07T20:52:44.000Z","title":"GHSA-8qf3-x8v5-2pj8 / CVE-2025-54368: uv allows ZIP payload obfuscation through parsing differentials","url":"https://github.com/advisories/GHSA-8qf3-x8v5-2pj8","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"181768417122","published":"2025-08-07T20:14:00.000Z","title":"CISA orders fed agencies to patch new Exchange flaw by Monday","url":"https://bleepingcomputer.com/news/security/cisa-orders-fed-agencies-to-patch-new-cve-2025-53786-exchange-flaw","summary":"CISA has issued an emergency directive ordering all Federal Civilian Executive Branch (FCEB) agencies to mitigate a critical Microsoft Exchange hybrid vulnerability tracked as CVE-2025-53786 by Monday morning at 9:00 AM ET.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"advisory"},{"id":"e41adc1d568a","published":"2025-08-07T17:58:00.000Z","title":"New EDR killer tool used by eight different ransomware groups","url":"https://bleepingcomputer.com/news/security/new-edr-killer-tool-used-by-eight-different-ransomware-groups","summary":"A new Endpoint Detection and Response (EDR) killer that is considered to be the evolution of 'EDRKillShifter,' developed by RansomHub, has been observed in attacks by eight different ransomware gangs.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"40c451707fe9","published":"2025-08-07T17:41:33.000Z","title":"Provable one-poison backdoor attacks on linear models and ReLU neural networks","url":"https://arxiv.org/abs/2508.05600","summary":"Backdoor poisoning attacks are a threat to machine learning models that are trained on data collected from untrusted sources; these attacks enable attackers to inject malicious behavior into the model that can be triggered by specially crafted inputs. Prior work has established…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"74691702e61d","published":"2025-08-07T17:34:41.000Z","title":"GHSA-93jv-pvg8-hf3v / CVE-2025-44779: Ollama allows deletion of arbitrary files","url":"https://github.com/advisories/GHSA-93jv-pvg8-hf3v","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d6b5418e67ef","published":"2025-08-07T16:42:54.000Z","title":"GHSA-378x-6p4f-8jgm / CVE-2025-54886: SKOPS Card.get_model happily allows arbitrary code execution","url":"https://github.com/advisories/GHSA-378x-6p4f-8jgm","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c09a34038b25","published":"2025-08-07T16:41:55.000Z","title":"GHSA-cq8c-xv66-36gw / CVE-2025-54793: Astros's duplicate trailing slash feature leads to an open redirection security issue","url":"https://github.com/advisories/GHSA-cq8c-xv66-36gw","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"cee3a4a32643","published":"2025-08-07T16:00:00.000Z","title":"How Hackers Exploit AI’s Problem-Solving Instincts","url":"https://developer.nvidia.com/blog/how-hackers-exploit-ais-problem-solving-instincts","summary":"As multimodal AI models advance from perception to reasoning, and even start acting autonomously, new attack surfaces emerge. These threats don’t just target inputs or outputs; they exploit how AI systems process, synthesize, and reason across modalities. The NVIDIA AI Red Team…","feed":"nvidia-dev","publisher":"nvidia","topics":[],"type":"analysis"},{"id":"027491553959","published":"2025-08-07T15:33:00.000Z","title":"Webinar: How to Stop Python Supply Chain Attacks—and the Expert Tools You Need","url":"https://thehackernews.com/2025/08/webinar-how-to-stop-python-supply-chain.html","summary":"Python is everywhere in modern software. From machine learning models to production microservices, chances are your code—and your business—depends on Python packages you didn’t write. But in 2025, that trust comes with a serious risk. Every few weeks, we’re seeing fresh…","feed":"thehackernews","publisher":"thehackernews","topics":["sca"],"type":"guide"},{"id":"37e9b96315a4","published":"2025-08-07T15:27:00.000Z","title":"SonicWall finds no SSLVPN zero-day, links ransomware attacks to 2024 flaw","url":"https://bleepingcomputer.com/news/security/sonicwall-finds-no-sslvpn-zero-day-links-ransomware-attacks-to-2024-flaw","summary":"SonicWall says that recent Akira ransomware attacks exploiting Gen 7 firewalls with SSLVPN enabled are exploiting an older vulnerability rather than a zero-day flaw.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"ab0ec233cf0d","published":"2025-08-07T15:20:58.000Z","title":"How Devin AI Can Leak Your Secrets via Multiple Means","url":"https://embracethered.com/blog/posts/2025/devin-can-leak-your-secrets","summary":"In this post we show how an attacker can make Devin send sensitive information to third-party servers, via multiple means. This post assumes that you read the first post about Devin as well. But here is a quick recap: During an indirect prompt injection Devin can be tricked…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["ai-security"],"type":"incident"},{"id":"d60811af0420","published":"2025-08-07T14:00:00.000Z","title":"Wave of 150 crypto-draining extensions hits Firefox add-on store","url":"https://bleepingcomputer.com/news/security/wave-of-150-crypto-draining-extensions-hits-firefox-add-on-store","summary":"A malicious campaign dubbed 'GreedyBear' has snuck onto the Mozilla add-ons store, targeting Firefox users with 150 malicious extensions and stealing an estimated $1,000,000 from unsuspecting victims.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"c09b66f113cf","published":"2025-08-07T13:56:54.000Z","title":"NT-ML: Backdoor Defense via Non-target Label Training and Mutual Learning","url":"https://arxiv.org/abs/2508.05404","summary":"Recent studies have shown that deep neural networks (DNNs) are vulnerable to backdoor attacks, where a designed trigger is injected into the dataset, causing erroneous predictions when activated. In this paper, we propose a novel defense mechanism, Non-target label Training and…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"2389e1525a25","published":"2025-08-07T13:19:00.002Z","title":"Malicious Go, npm Packages Deliver Cross-Platform Malware, Trigger Remote Data Wipes","url":"https://thehackernews.com/2025/08/malicious-go-npm-packages-deliver-cross.html","summary":"Cybersecurity researchers have discovered a set of 11 malicious Go packages that are designed to download additional payloads from remote servers and execute them on both Windows and Linux systems. \"At runtime the code silently spawns a shell, pulls a second-stage payload from…","feed":"thehackernews","publisher":"thehackernews","topics":["sca"],"type":"research"},{"id":"be0f02dfb5f8","published":"2025-08-07T11:00:02.000Z","title":"Here’s how deepfake vishing attacks work, and why they can be hard to detect","url":"https://arstechnica.com/security/2025/08/heres-how-deepfake-vishing-attacks-work-and-why-they-can-be-hard-to-detect","summary":"Why AI-based voice cloning is the next frontier in social-engineering attacks.","feed":"ars-security","publisher":"ars-security","topics":[],"type":"analysis"},{"id":"9c8745c24c19","published":"2025-08-07T10:45:00.000Z","title":"The AI-Powered Security Shift: What 2025 Is Teaching Us About Cloud Defense","url":"https://thehackernews.com/2025/08/the-ai-powered-security-shift-what-2025.html","summary":"Now that we are well into 2025, cloud attacks are evolving faster than ever and artificial intelligence (AI) is both a weapon and a shield. As AI rapidly changes how enterprises innovate, security teams are now tasked with a triple burden: Secure AI embedded in every part of…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"f2ab8ad82e88","published":"2025-08-07T10:42:00.003Z","title":"Microsoft Discloses Exchange Server Flaw Enabling Silent Cloud Access in Hybrid Setups","url":"https://thehackernews.com/2025/08/microsoft-discloses-exchange-server.html","summary":"Microsoft has released an advisory for a high-severity security flaw affecting on-premise versions of Exchange Server that could allow an attacker to gain elevated privileges under certain conditions. The vulnerability, tracked as CVE-2025-53786, carries a CVSS score of 8.0.…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"9bcef0ed17fd","published":"2025-08-07T10:32:00.003Z","title":"SonicWall Confirms Patched Vulnerability Behind Recent VPN Attacks, Not a Zero-Day","url":"https://thehackernews.com/2025/08/sonicwall-confirms-patched.html","summary":"SonicWall has revealed that the recent spike in activity targeting its Gen 7 and newer firewalls with SSL VPN enabled is related to an older, now-patched bug and password reuse. \"We now have high confidence that the recent SSL VPN activity is not connected to a zero-day…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"advisory"},{"id":"ef00107acefb","published":"2025-08-07T07:50:00.000Z","title":"Microsoft warns of high-severity flaw in hybrid Exchange deployments","url":"https://bleepingcomputer.com/news/microsoft/microsoft-warns-of-high-severity-flaw-in-hybrid-exchange-deployments","summary":"Microsoft has warned customers to mitigate a high-severity vulnerability in Exchange Server hybrid deployments that could allow attackers to escalate privileges in Exchange Online cloud environments undetected.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"9771f39a8105","published":"2025-08-07T07:14:01.000Z","title":"JPS: Jailbreak Multimodal Large Language Models with Collaborative Visual Perturbation and Textual Steering","url":"https://arxiv.org/abs/2508.05087","summary":"Jailbreak attacks against multimodal large language Models (MLLMs) are a significant research focus. Current research predominantly focuses on maximizing attack success rate (ASR), often overlooking whether the generated responses actually fulfill the attacker's malicious…","feed":"arxiv-agentic","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"aec54e92a7f3","published":"2025-08-07T00:00:00.000Z","title":"From hard refusals to safe-completions: toward output-centric safety training","url":"https://openai.com/index/gpt-5-safe-completions","summary":"Discover how OpenAI's new safe-completions approach in GPT-5 improves both safety and helpfulness in AI responses—moving beyond hard refusals to nuanced, output-centric safety training for handling dual-use prompts.","feed":"openai","publisher":"openai","topics":["models"],"type":"analysis"},{"id":"ec7686769469","published":"2025-08-07T00:00:00.000Z","title":"Preventing ZIP parser confusion attacks on Python package installers","url":"https://blog.pypi.org/posts/2025-08-07-wheel-archive-confusion-attacks","summary":"PyPI will begin warning and will later reject wheels that contain differentiable ZIP features or incorrect RECORD files.","feed":"pypi-blog","publisher":"pypi","topics":[],"type":"incident"},{"id":"f26617f729ec","published":"2025-08-06T20:30:00.002Z","title":"Researchers Uncover ECScape Flaw in Amazon ECS Enabling Cross-Task Credential Theft","url":"https://thehackernews.com/2025/08/researchers-uncover-ecscape-flaw-in.html","summary":"Cybersecurity researchers have demonstrated an \"end-to-end privilege escalation chain\" in Amazon Elastic Container Service (ECS) that could be exploited by an attacker to conduct lateral movement, access sensitive data, and seize control of the cloud environment. The attack…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"research"},{"id":"a87926cb79e8","published":"2025-08-06T20:00:00.003Z","title":"Fake VPN and Spam Blocker Apps Tied to VexTrio Used in Ad Fraud, Subscription Scams","url":"https://thehackernews.com/2025/08/fake-vpn-and-spam-blocker-apps-tied-to.html","summary":"The malicious ad tech purveyor known as VexTrio Viper has been observed developing several malicious apps that have been published on Apple and Google's official app storefronts under the guise of seemingly useful applications. These apps masquerade as VPNs, device \"monitoring\"…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"e64066af2ca5","published":"2025-08-06T19:40:00.000Z","title":"TeaOnHer, a rival Tea app for men, is leaking users’ personal data and driver’s licenses","url":"https://techcrunch.com/2025/08/06/a-rival-tea-app-for-men-is-leaking-its-users-personal-data-and-drivers-licenses","summary":"The newly launched app, now trending on Apple's App Store, contains at least one major security flaw that exposes the private information of its users, including their uploaded selfies and government-issued IDs.","feed":"techcrunch-security","publisher":"techcrunch-security","topics":[],"type":"analysis"},{"id":"c53d5e91a645","published":"2025-08-06T17:11:42.000Z","title":"GHSA-8q6v-474h-whgg / CVE-2025-54885: The Thinbus Javascript Secure Remote Password (SRP) Client Generates Fewer Bits of Entropy Than Intended","url":"https://github.com/advisories/GHSA-8q6v-474h-whgg","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"28e6de5564cf","published":"2025-08-06T17:08:52.000Z","title":"GHSA-q82r-2j7m-9rv4 / CVE-2025-54799: github.com/go-acme/lego/v4/acme/api does not enforce HTTPS","url":"https://github.com/advisories/GHSA-q82r-2j7m-9rv4","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"abc267dacc8f","published":"2025-08-06T17:06:04.000Z","title":"GHSA-52f5-9888-hmc6 / CVE-2025-54798: tmp allows arbitrary temporary file / directory write via symbolic link `dir` parameter","url":"https://github.com/advisories/GHSA-52f5-9888-hmc6","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"2000533d9449","published":"2025-08-06T16:36:00.000Z","title":"New Ghost Calls tactic abuses Zoom and Microsoft Teams for C2 operations","url":"https://bleepingcomputer.com/news/security/new-ghost-calls-tactic-abuses-zoom-and-microsoft-teams-for-c2-operations","summary":"A new post-exploitation command-and-control (C2) evasion method called 'Ghost Calls' abuses TURN servers used by conferencing apps like Zoom and Microsoft Teams to tunnel traffic through trusted infrastructure.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"459c5285060a","published":"2025-08-06T16:00:00.000Z","title":"Sharing practical guidance: Launching Microsoft Secure Future Initiative (SFI) patterns and practices","url":"https://microsoft.com/en-us/security/blog/2025/08/06/sharing-practical-guidance-launching-microsoft-secure-future-initiative-sfi-patterns-and-practices","summary":"We’re excited to launch SFI patterns and practices: a new library of actionable guidance designed to help organizations implement security measures at scale. This launch marks a next step in our journey to make our SFI learnings practical for our customers, partners, and…","feed":"microsoft-security","publisher":"microsoft","topics":[],"type":"standard"},{"id":"23aee289fe9e","published":"2025-08-06T14:18:08.000Z","title":"OpenSSF at DEF CON 33: AI Cyber Challenge (AIxCC), MLSecOps, and Securing Critical Infrastructure","url":"https://openssf.org/blog/2025/08/06/openssf-at-def-con-33-ai-cyber-challenge-aixcc-mlsecops-and-securing-critical-infrastructure","summary":"By Jeff Diecks The OpenSSF team will be attending DEF CON 33 , where the winners of the AI Cyber Challenge (AIxCC) will be announced. We will also host a panel discussion at the AIxCC village to introduce the concept of MLSecOps. AIxCC, led by DARPA and ARPA-H , is a two-year…","feed":"openssf","publisher":"openssf","topics":["sast","remediation"],"type":"analysis"},{"id":"dd944b1b087d","published":"2025-08-06T13:55:43.000Z","title":"Empirical Evaluation of AI-Assisted Software Package Selection: A Knowledge Graph Approach","url":"https://arxiv.org/abs/2508.05693","summary":"Selecting third-party software packages in open-source ecosystems like Python is challenging due to the large number of alternatives and limited transparent evidence for comparison. Generative AI tools are increasingly used in development workflows, but their suggestions often…","feed":"arxiv-sca","publisher":"arxiv","topics":["sca"],"type":"research"},{"id":"9de4119f6604","published":"2025-08-06T13:48:38.000Z","title":"Large Language Models Versus Static Code Analysis Tools: A Systematic Benchmark for Vulnerability Detection","url":"https://arxiv.org/abs/2508.04448","summary":"Modern software relies on a multitude of automated testing and quality assurance tools to prevent errors, bugs and potential vulnerabilities. This study sets out to provide a head-to-head, quantitative and qualitative evaluation of six automated approaches: three…","feed":"arxiv-sca","publisher":"arxiv","topics":["sast","sca"],"type":"research"},{"id":"44756f8bee81","published":"2025-08-06T12:31:20.000Z","title":"GHSA-7rx2-769v-hrwf / CVE-2025-6013: HashiCorp Vault ldap auth method may not have correctly enforced MFA","url":"https://github.com/advisories/GHSA-7rx2-769v-hrwf","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"eb5d566daddb","published":"2025-08-06T12:31:20.000Z","title":"GHSA-9356-575x-2w9m / CVE-2025-5197: Hugging Face Transformers Regular Expression Denial of Service (ReDoS) vulnerability","url":"https://github.com/advisories/GHSA-9356-575x-2w9m","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"558107bc8f99","published":"2025-08-06T12:05:25.000Z","title":"Google says hackers stole its customers’ data by breaching its Salesforce database","url":"https://techcrunch.com/2025/08/06/google-says-hackers-stole-its-customers-data-in-a-breach-of-its-salesforce-database","summary":"Google attributed the breach to a hacking group, ShinyHunters, known for targeting big companies and their cloud-based databases.","feed":"techcrunch-security","publisher":"techcrunch-security","topics":[],"type":"incident"},{"id":"5acbb295d915","published":"2025-08-06T11:58:00.000Z","title":"ReVault flaws let hackers bypass Windows login on Dell laptops","url":"https://bleepingcomputer.com/news/security/revault-flaws-let-hackers-bypass-windows-login-on-dell-laptops","summary":"ControlVault3 firmware vulnerabilities impacting over 100 Dell laptop models can allow attackers to bypass Windows login and install malware that persists across system reinstalls.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"1c14b68ed486","published":"2025-08-06T10:36:00.001Z","title":"Microsoft Launches Project Ire to Autonomously Classify Malware Using AI Tools","url":"https://thehackernews.com/2025/08/microsoft-launches-project-ire-to.html","summary":"Microsoft on Tuesday announced an autonomous artificial intelligence (AI) agent that can analyze and classify software without assistance in an effort to advance malware detection efforts. The large language model (LLM)-powered autonomous malware classification system,…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"announcement"},{"id":"aed24f68124d","published":"2025-08-06T10:10:01.000Z","title":"Safeguarding Deliberation in Digital Democracy","url":"https://arxiv.org/abs/2508.04281","summary":"Large Language Models (LLMs) are gaining traction in digital democracy experiments as a method to aggregate preferences through deliberation and express the resulting common ground as consensus statements. Yet, participants can exploit critical vulnerabilities in such LLM-based…","feed":"arxiv-aisec","publisher":"arxiv","topics":[],"type":"research"},{"id":"a3faed0f41e7","published":"2025-08-06T10:06:00.000Z","title":"Trend Micro warns of Apex One zero-day exploited in attacks","url":"https://bleepingcomputer.com/news/security/trend-micro-warns-of-endpoint-protection-zero-day-exploited-in-attacks","summary":"Trend Micro has warned customers to immediately secure their systems against an actively exploited remote code execution vulnerability in its Apex One endpoint security platform.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"9c5ebc37d31f","published":"2025-08-06T10:00:47.000Z","title":"When Good Accounts Go Bad: Exploiting Delegated Managed Service Accounts in Active Directory","url":"https://unit42.paloaltonetworks.com/badsuccessor-attack-vector","summary":"BadSuccessor is an attack vector in Windows Server 2025. Under certain conditions it allows privilege elevation via dMSAs. We analyze its mechanics.","feed":"unit42","publisher":"palo-alto","topics":[],"type":"analysis"},{"id":"8647b71c11c1","published":"2025-08-06T09:30:37.000Z","title":"GHSA-27gv-mg7w-mm34 / CVE-2025-7954: Shopware race condition bypasses voucher restrictions","url":"https://github.com/advisories/GHSA-27gv-mg7w-mm34","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"0d61a55b2e28","published":"2025-08-06T08:57:00.003Z","title":"Trend Micro Confirms Active Exploitation of Critical Apex One Flaws in On-Premise Systems","url":"https://thehackernews.com/2025/08/trend-micro-confirms-active.html","summary":"Trend Micro has released mitigations to address critical security flaws in on-premise versions of Apex One Management Console that it said have been exploited in the wild. The vulnerabilities (CVE-2025-54948 and CVE-2025-54987), both rated 9.4 on the CVSS scoring system, have…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"analysis"},{"id":"d90af2692d1a","published":"2025-08-06T08:48:55.000Z","title":"Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction","url":"https://arxiv.org/abs/2508.04216","summary":"External reasoning systems combine language models with process reward models (PRMs) to select high-quality reasoning paths for complex tasks such as mathematical problem solving. However, these systems are prone to reward hacking, where high-scoring but logically incorrect…","feed":"arxiv-models","publisher":"arxiv","topics":["trust","models"],"type":"research"},{"id":"fdf135c42f87","published":"2025-08-06T08:25:40.000Z","title":"Eliciting and Analyzing Emergent Misalignment in State-of-the-Art Large Language Models","url":"https://arxiv.org/abs/2508.04196","summary":"Despite significant advances in alignment techniques, we demonstrate that state-of-the-art language models remain vulnerable to carefully crafted conversational scenarios that can induce various forms of misalignment without explicit jailbreaking. Through systematic manual…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"6bea15efae2d","published":"2025-08-06T08:18:01.000Z","title":"BadTime: An Effective Backdoor Attack on Multivariate Long-Term Time Series Forecasting","url":"https://arxiv.org/abs/2508.04189","summary":"Multivariate long-term time series forecasting (MLTSF) models are increasingly deployed in critical domains such as climate, finance, and transportation. Despite their growing importance, the security of MLTSF models against backdoor attacks remains entirely unexplored. To…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"ab84f51b84d3","published":"2025-08-06T08:01:58.000Z","title":"I Spent $500 To Test Devin AI For Prompt Injection So That You Don't Have To","url":"https://embracethered.com/blog/posts/2025/devin-i-spent-usd500-to-hack-devin","summary":"Today we cover Devin AI from Cognition, the first AI Software Engineer. We will cover Devin proof-of-concept exploits in multiple posts over the next few days. In this first post, we show how a prompt injection payload hosted on a website leads to a full compromise of Devin’s…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","ai-security"],"type":"research"},{"id":"16fce90caf4e","published":"2025-08-06T05:50:00.003Z","title":"AI Is Transforming Cybersecurity Adversarial Testing - Pentera Founder’s Vision","url":"https://thehackernews.com/2025/08/ai-is-transforming-cybersecurity.html","summary":"When Technology Resets the Playing Field In 2015 I founded a cybersecurity testing software company with the belief that automated penetration testing was not only possible, but necessary. At the time, the idea was often met with skepticism, but today, with 1200+ of enterprise…","feed":"thehackernews","publisher":"thehackernews","topics":["dast"],"type":"analysis"},{"id":"93d5cf410230","published":"2025-08-06T05:24:55.000Z","title":"Risky Business #801 -- AI models can hack well now and it's weirding us out","url":"https://risky.biz/RB801","summary":"On this week’s show Patrick Gray and Adam Boileau discuss the week’s cybersecurity news. Google security engineering VP Heather Adkins drops by to talk about their AI bug hunter, and Risky Business producer Amberleigh Jack makes her main show debut. This episode explores the…","feed":"riskybiz","publisher":"riskybiz","topics":[],"type":"roundup"},{"id":"de518585ba36","published":"2025-08-06T05:21:40.000Z","title":"Isolate Trigger: Detecting and Eliminating Adaptive Backdoor Attacks","url":"https://arxiv.org/abs/2508.04094","summary":"Deep learning models are widely deployed in various applications but remain vulnerable to stealthy adversarial threats, particularly backdoor attacks. Backdoor models trained on poisoned datasets behave normally with clean inputs but cause mispredictions when a specific trigger…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"36f2728924f2","published":"2025-08-06T03:54:29.000Z","title":"FLAT: Revealing Hidden Latent-Conditioned Backdoor Failures in Federated Learning","url":"https://arxiv.org/abs/2508.04064","summary":"Horizontal federated learning (HFL) backdoor audits often summarize model behavior through clean accuracy (CA), mean attack success rate (ASR), or a single known-trigger test. Such summaries can hide a different failure mode, in which one target label is activated by many…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"5837b397dd18","published":"2025-08-06T00:30:25.000Z","title":"GHSA-4pcg-pjp5-3mc6 / CVE-2025-8571: Concrete CMS vulnerable to Reflected Cross-Site Scripting (XSS) in Conversation Messages Dashboard Page","url":"https://github.com/advisories/GHSA-4pcg-pjp5-3mc6","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a00b58fb49a3","published":"2025-08-06T00:30:25.000Z","title":"GHSA-c5xf-rmv4-j85h / CVE-2025-8573: Concrete CMS is vulnerable to Stored XSS from Home Folder on Members Dashboard page","url":"https://github.com/advisories/GHSA-c5xf-rmv4-j85h","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"356e1ca0d0c1","published":"2025-08-05T23:30:20.000Z","title":"SonicWall firewalls hit by active mass exploitation of suspected zero-day","url":"https://cyberscoop.com/sonicwall-firewalls-attack-spree-zero-day","summary":"About 20 organizations have been impacted and the pace of attacks is rising. Threat researchers and SonicWall are scrambling to determine the root cause.","feed":"cyberscoop","publisher":"cyberscoop","topics":[],"type":"research"},{"id":"40d97702d099","published":"2025-08-05T21:57:52.000Z","title":"ASTRA: Autonomous Spatial-Temporal Red-teaming for AI Software Assistants","url":"https://arxiv.org/abs/2508.03936","summary":"AI coding assistants like GitHub Copilot are rapidly transforming software development, but their safety remains deeply uncertain-especially in high-stakes domains like cybersecurity. Current red-teaming tools often rely on fixed benchmarks or unrealistic prompts, missing many…","feed":"arxiv-agentic","publisher":"arxiv","topics":["agentic-sdlc"],"type":"research"},{"id":"13589022d0f8","published":"2025-08-05T21:06:00.000Z","title":"PBS confirms data breach after employee info leaked on Discord servers","url":"https://bleepingcomputer.com/news/security/pbs-confirms-data-breach-after-employee-info-leaked-on-discord-servers","summary":"PBS has suffered a data breach exposing the corporate contact information of its employees and those of its affiliates, BleepingComputer has learned.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"2a1362cb7372","published":"2025-08-05T19:43:25.000Z","title":"RX-INT: A Kernel Engine for Real-Time Detection and Analysis of In-Memory Threats","url":"https://arxiv.org/abs/2508.03879","summary":"Malware and cheat developers use fileless execution techniques to evade traditional, signature-based security products. These methods include various types of manual mapping, module stomping, and threadless injection which work entirely within the address space of a legitimate…","feed":"arxiv-sca","publisher":"arxiv","topics":["sca"],"type":"research"},{"id":"d35d0d0f3dca","published":"2025-08-05T19:14:37.000Z","title":"Evaluating Software Supply Chain Security in Research Software","url":"https://arxiv.org/abs/2508.03856","summary":"The security of research software is essential for ensuring the integrity and reproducibility of scientific results. However, research software security is still largely unexplored. Due to its dependence on open source components and distributed development practices, research…","feed":"arxiv-sca","publisher":"arxiv","topics":["sca"],"type":"research"},{"id":"4484a67b16e6","published":"2025-08-05T17:56:51.000Z","title":"AttnTrace: Contextual Attribution of Prompt Injection and Knowledge Corruption","url":"https://arxiv.org/abs/2508.03793","summary":"Long-context large language models (LLMs), such as Gemini-2.5-Pro and Claude-Sonnet-4, are increasingly used to empower advanced AI systems, including retrieval-augmented generation (RAG) pipelines and autonomous agents. In these systems, an LLM receives an instruction along…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"4c2c7ca287d9","published":"2025-08-05T17:42:06.000Z","title":"GHSA-f6rc-24x4-ppxp / CVE-2025-54873: RISC Zero Underconstrained Vulnerability: Division","url":"https://github.com/advisories/GHSA-f6rc-24x4-ppxp","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ff71a116632e","published":"2025-08-05T17:13:24.000Z","title":"GHSA-57q2-6cp4-9mq3 / CVE-2025-54125: XWiki exposes passwords and emails stored in fields not named password/email in xml.vm","url":"https://github.com/advisories/GHSA-57q2-6cp4-9mq3","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6c2b9b6cdd56","published":"2025-08-05T17:12:03.000Z","title":"GHSA-r38m-cgpg-qj69 / CVE-2025-54124: XWiki leaks password hashes and other accessible password properties","url":"https://github.com/advisories/GHSA-r38m-cgpg-qj69","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"72cf2455f566","published":"2025-08-05T16:17:02.000Z","title":"ReFuzzer: Feedback-Driven Approach to Enhance Validity of LLM-Generated Test Programs","url":"https://arxiv.org/abs/2508.03603","summary":"Existing LLM-based compiler fuzzers often produce syntactically or semantically invalid test programs, limiting their effectiveness in exercising compiler optimizations and backend components. We introduce ReFuzzer, a framework for refining LLM-generated test programs by…","feed":"arxiv-dast","publisher":"arxiv","topics":["dast"],"type":"research"},{"id":"c1de2f1501c4","published":"2025-08-05T15:59:19.000Z","title":"Selection-Based Vulnerabilities: Clean-Label Backdoor Attacks in Active Learning","url":"https://arxiv.org/abs/2508.05681","summary":"Active learning(AL), which serves as the representative label-efficient learning paradigm, has been widely applied in resource-constrained scenarios. The achievement of AL is attributed to acquisition functions, which are designed for identifying the most important data to…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"ce9567a1ef0b","published":"2025-08-05T15:54:52.000Z","title":"MalFlows: Context-aware Fusion of Heterogeneous Flow Semantics for Android Malware Detection","url":"https://arxiv.org/abs/2508.03588","summary":"Static analysis, a fundamental technique in Android app examination, enables the extraction of control flows, data flows, and inter-component communications (ICCs), all of which are essential for malware detection. However, existing methods struggle to leverage the semantic…","feed":"arxiv-crse","publisher":"arxiv","topics":[],"type":"research"},{"id":"8ccc9377c46c","published":"2025-08-05T15:32:56.000Z","title":"GHSA-m9x4-w7p9-mxhx / CVE-2025-32430: XWiki allows Reflected XSS in two templates","url":"https://github.com/advisories/GHSA-m9x4-w7p9-mxhx","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"f0e0d04292a7","published":"2025-08-05T15:30:53.000Z","title":"GHSA-mrwc-mvr8-9xq5 / CVE-2025-50706: ThinkPHP Path Traversal Vulnerability","url":"https://github.com/advisories/GHSA-mrwc-mvr8-9xq5","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"682e87f12b6e","published":"2025-08-05T15:23:54.000Z","title":"GHSA-jxhh-4648-vpp3 / CVE-2025-54869: FPDI allows Memory Exhaustion (OOM) in PDF Parser which leads to Denial of Service","url":"https://github.com/advisories/GHSA-jxhh-4648-vpp3","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d696923cf0af","published":"2025-08-05T15:22:21.000Z","title":"GHSA-qx2q-88mx-vhg7 / CVE-2025-54801: Fiber Crashes in BodyParser Due to Unvalidated Large Slice Index in Decoder","url":"https://github.com/advisories/GHSA-qx2q-88mx-vhg7","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c554fe11f63c","published":"2025-08-05T14:13:34.000Z","title":"Visualizing Secure MLOps (MLSecOps): A Practical Guide for Building Robust AI/ML Pipeline Security","url":"https://openssf.org/blog/2025/08/05/visualizing-secure-mlops-mlsecops-a-practical-guide-for-building-robust-ai-ml-pipeline-security","summary":"By Sarah Evans and Andrey Shorov The world of technology is constantly evolving, and with the rise of Artificial Intelligence (AI) and Machine Learning (ML), the demand for robust security...","feed":"openssf","publisher":"openssf","topics":[],"type":"guide"},{"id":"7dc0e3b416b5","published":"2025-08-05T14:12:33.000Z","title":"GHSA-vf9j-h32g-2764 / CVE-2025-54073: mcp-package-docs vulnerable to command injection in several tools","url":"https://github.com/advisories/GHSA-vf9j-h32g-2764","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca","ai-security"],"type":"advisory"},{"id":"f75a277d8041","published":"2025-08-05T14:05:32.000Z","title":"On the Evaluation of Large Language Models in Multilingual Vulnerability Repair","url":"https://arxiv.org/abs/2508.03470","summary":"Various Deep Learning-based approaches with pre-trained language models have been proposed for automatically repairing software vulnerabilities. However, these approaches are limited to a specific programming language (C/C++). Recent advances in large language models (LLMs)…","feed":"arxiv-remediation","publisher":"arxiv","topics":["remediation"],"type":"research"},{"id":"d84a420daa43","published":"2025-08-05T13:59:00.003Z","title":"Google’s August Patch Fixes Two Qualcomm Vulnerabilities Exploited in the Wild","url":"https://thehackernews.com/2025/08/google-fixes-3-android-vulnerabilities.html","summary":"Google has released security updates to address multiple security flaws in Android, including fixes for two Qualcomm bugs that were flagged as actively exploited in the wild. The vulnerabilities include CVE-2025-21479 (CVSS score: 8.6) and CVE-2025-27038 (CVSS score: 7.5), both…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"incident"},{"id":"7baa3be27ba7","published":"2025-08-05T13:48:43.000Z","title":"SonicWall urges customers to disable SSLVPN amid reports of ransomware attacks","url":"https://techcrunch.com/2025/08/05/sonicwall-urges-customers-to-disable-sslvpn-amid-reports-of-ransomware-attacks","summary":"Security researchers say they have evidence that ransomware gangs are hacking into large companies that rely on fully patched SonicWall firewalls. The researchers say it's likely the flaw is a \"zero-day\" bug currently unknown to SonicWall.","feed":"techcrunch-security","publisher":"techcrunch-security","topics":[],"type":"incident"},{"id":"486dc37bd220","published":"2025-08-05T13:20:58.000Z","title":"Amp Code: Arbitrary Command Execution via Prompt Injection Fixed","url":"https://embracethered.com/blog/posts/2025/amp-agents-that-modify-system-configuration-and-escape","summary":"Sandbox-escape-style attacks can happen when an AI is able to modify its own configuration settings, such as by writing to configuration files. That was the case with Amp , an agentic coding tool built by Sourcegraph . The AI coding agent could update its own configuration and:…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc","ai-security"],"type":"analysis"},{"id":"ebed3af361a0","published":"2025-08-05T13:01:00.001Z","title":"Cursor AI Code Editor Vulnerability Enables RCE via Malicious MCP File Swaps Post Approval","url":"https://thehackernews.com/2025/08/cursor-ai-code-editor-vulnerability.html","summary":"Cybersecurity researchers have disclosed a high-severity security flaw in the artificial intelligence (AI)-powered code editor Cursor that could result in remote code execution. The vulnerability, tracked as CVE-2025-54136 (CVSS score: 7.2), has been codenamed MCPoison by Check…","feed":"thehackernews","publisher":"thehackernews","topics":["ai-security"],"type":"research"},{"id":"8862d48f5e38","published":"2025-08-05T12:14:01.000Z","title":"When Good Sounds Go Adversarial: Jailbreaking Audio-Language Models with Benign Inputs","url":"https://arxiv.org/abs/2508.03365","summary":"As large language models (LLMs) become increasingly integrated into daily life, audio has emerged as a key interface for human-AI interaction. However, this convenience also introduces new vulnerabilities, making audio a potential attack surface for adversaries. Our research…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"813439c07616","published":"2025-08-05T11:28:00.000Z","title":"SonicWall urges admins to disable SSLVPN amid rising attacks","url":"https://bleepingcomputer.com/news/security/sonicwall-urges-admins-to-disable-sslvpn-amid-rising-attacks","summary":"SonicWall has warned customers to disable SSLVPN services due to ransomware gangs potentially exploiting an unknown security vulnerability in SonicWall Gen 7 firewalls to breach networks over the past few weeks.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"316680dbce0f","published":"2025-08-05T11:25:00.001Z","title":"Misconfigurations Are Not Vulnerabilities: The Costly Confusion Behind Security Risks","url":"https://thehackernews.com/2025/08/misconfigurations-are-not.html","summary":"In SaaS security conversations, “misconfiguration” and “vulnerability” are often used interchangeably. But they’re not the same thing. And misunderstanding that distinction can quietly create real exposure. This confusion isn’t just semantics. It reflects a deeper…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"f1c80d2f4529","published":"2025-08-05T10:31:00.000Z","title":"Android gets patches for Qualcomm flaws exploited in attacks","url":"https://bleepingcomputer.com/news/security/android-gets-patches-for-qualcomm-flaws-exploited-in-attacks","summary":"Google has released security patches for six vulnerabilities in Android's August 2025 security update, including two Qualcomm flaws exploited in targeted attacks.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["remediation"],"type":"advisory"},{"id":"8d57a3a8676c","published":"2025-08-05T10:28:07.000Z","title":"BDFirewall: Towards Effective and Expeditiously Black-Box Backdoor Defense in MLaaS","url":"https://arxiv.org/abs/2508.03307","summary":"In this paper, we endeavor to address the challenges of backdoor attacks countermeasures in black-box scenarios, thereby fortifying the security of inference under MLaaS. We first categorize backdoor triggers from a new perspective, i.e., their impact on the patched area, and…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"37320b475bf7","published":"2025-08-05T08:48:37.000Z","title":"BadBlocks: Low-Cost and Stealthy Backdoor Attacks Tailored for Text-to-Image Diffusion Models","url":"https://arxiv.org/abs/2508.03221","summary":"Despite the remarkable progress of diffusion models in image generation, recent studies reveal their vulnerability to backdoor attacks via covert visual or textual triggers. Although evolving defense mechanisms can detect most existing threats through visual inspection or…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"8439ba2f85b4","published":"2025-08-05T07:44:33.000Z","title":"Trustworthiness of Legal Considerations for the Use of LLMs in Education","url":"https://arxiv.org/abs/2508.03771","summary":"As Artificial Intelligence (AI), particularly Large Language Models (LLMs), becomes increasingly embedded in education systems worldwide, ensuring their ethical, legal, and contextually appropriate deployment has become a critical policy concern. This paper offers a comparative…","feed":"arxiv-trust","publisher":"arxiv","topics":["trust"],"type":"research"},{"id":"ca0a466d2a17","published":"2025-08-05T07:27:00.003Z","title":"15,000 Fake TikTok Shop Domains Deliver Malware, Steal Crypto via AI-Driven Scam Campaign","url":"https://thehackernews.com/2025/08/15000-fake-tiktok-shop-domains-deliver.html","summary":"Cybersecurity researchers have lifted the veil on a widespread malicious campaign that's targeting TikTok Shop users globally with an aim to steal credentials and distribute trojanized apps. \"Threat actors are exploiting the official in-app e-commerce platform through a dual…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"incident"},{"id":"fc2c7f7635fd","published":"2025-08-05T06:14:53.000Z","title":"Attack the Messages, Not the Agents: A Multi-round Adaptive Stealthy Tampering Framework for LLM-MAS","url":"https://arxiv.org/abs/2508.03125","summary":"Large language model-based multi-agent systems (LLM-MAS) effectively accomplish complex and dynamic tasks through inter-agent communication, but this reliance introduces substantial safety vulnerabilities. Existing attack methods targeting LLM-MAS either compromise agent…","feed":"arxiv-agentic","publisher":"arxiv","topics":[],"type":"research"},{"id":"562d990ea075","published":"2025-08-05T05:27:27.000Z","title":"Using the NANDA Index Architecture in Practice: An Enterprise Perspective","url":"https://arxiv.org/abs/2508.03101","summary":"The proliferation of autonomous AI agents represents a paradigmatic shift from traditional web architectures toward collaborative intelligent systems requiring sophisticated mechanisms for discovery, authentication, capability verification, and secure collaboration across…","feed":"arxiv-aisec","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"5b05eee8a64c","published":"2025-08-05T05:18:00.004Z","title":"SonicWall Investigating Potential SSL VPN Zero-Day After 20+ Targeted Attacks Reported","url":"https://thehackernews.com/2025/08/sonicwall-investigating-potential-ssl.html","summary":"SonicWall said it's actively investigating reports to determine if there is a new zero-day vulnerability following reports of a spike in Akira ransomware actors in late July 2025. \"Over the past 72 hours, there has been a notable increase in both internally and externally…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"175756bbfb99","published":"2025-08-05T05:17:18.000Z","title":"Evolution of AI Agent Registry Solutions: Centralized, Enterprise, and Distributed Approaches","url":"https://arxiv.org/abs/2508.03095","summary":"Autonomous AI agents now operate across cloud, enterprise, and decentralized domains, creating demand for registry infrastructures that enable trustworthy discovery, capability negotiation, and identity assurance. We analyze five prominent approaches: (1) MCP Registry…","feed":"arxiv-aisec","publisher":"arxiv","topics":["trust","ai-security"],"type":"research"},{"id":"dfcbccba3e70","published":"2025-08-05T03:25:09.000Z","title":"Towards Effective Offensive Security LLM Agents: Hyperparameter Tuning, LLM as a Judge, and a Lightweight CTF Benchmark","url":"https://arxiv.org/abs/2508.05674","summary":"Recent advances in LLM agentic systems have improved the automation of offensive security tasks, particularly for Capture the Flag (CTF) challenges. We systematically investigate the key factors that drive agent success and provide a detailed recipe for building effective…","feed":"arxiv-agentic","publisher":"arxiv","topics":["dast"],"type":"research"},{"id":"6f4348fd25ae","published":"2025-08-05T00:30:26.000Z","title":"GHSA-3j6h-5v68-hvqg / CVE-2025-4604: Liferay Portal CAPTCHA Bypass for Gogo Shell","url":"https://github.com/advisories/GHSA-3j6h-5v68-hvqg","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"40d828f06347","published":"2025-08-05T00:00:00.000Z","title":"CVE-2022-40799: D-Link DNR-322L Download of Code Without Integrity Check Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2022-40799","summary":"D-Link DNR-322L contains a download of code without integrity check vulnerability that could allow an authenticated attacker to execute OS level commands on the device. The impacted products could be end-of-life (EoL) and/or end-of-service (EoS). Users should discontinue product ","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"944c7faf292d","published":"2025-08-05T00:00:00.000Z","title":"CVE-2020-25078: D-Link DCS-2530L and DCS-2670L Devices Unspecified Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2020-25078","summary":"D-Link DCS-2530L and DCS-2670L devices contains an unspecified vulnerability that could allow for remote administrator password disclosure. The impacted products could be end-of-life (EoL) and/or end-of-service (EoS). Users should discontinue product utilization. US federal agenc","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"eb18cea51b5c","published":"2025-08-05T00:00:00.000Z","title":"CVE-2020-25079: D-Link DCS-2530L and DCS-2670L Command Injection Vulnerability added to CISA's Known Exploited Vulnerabilities catalog","url":"https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2020-25079","summary":"D-Link DCS-2530L and DCS-2670L devices contains a command injection vulnerability in the cgi-bin/ddns_enc.cgi. The impacted products could be end-of-life (EoL) and/or end-of-service (EoS). Users should discontinue product utilization. US federal agencies must act by 2025-08-26.","feed":"cisa-kev","publisher":"cisa","topics":[],"type":"advisory"},{"id":"ec9ec81ea46b","published":"2025-08-04T23:52:15.000Z","title":"Defend LLMs Through Self-Consciousness","url":"https://arxiv.org/abs/2508.02961","summary":"This paper introduces a novel self-consciousness defense mechanism for Large Language Models (LLMs) to combat prompt injection attacks. Unlike traditional approaches that rely on external classifiers, our method leverages the LLM's inherent reasoning capabilities to perform…","feed":"arxiv-aisec","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"26d469b5dd4e","published":"2025-08-04T21:52:50.000Z","title":"PentestJudge: Judging Agent Behavior Against Operational Requirements","url":"https://arxiv.org/abs/2508.02921","summary":"We introduce PentestJudge, a system for evaluating the operations of penetration testing agents. PentestJudge is a large language model (LLM)-as-judge with access to tools that allow it to consume arbitrary trajectories of agent states and tool call history to determine whether…","feed":"arxiv-dast","publisher":"arxiv","topics":["dast"],"type":"research"},{"id":"f2f2dd3851f9","published":"2025-08-04T21:31:46.000Z","title":"Google addresses six vulnerabilities in August’s Android security update","url":"https://cyberscoop.com/android-security-update-august-2025","summary":"Android partners and customers have experienced a temporary respite from double-digit vulnerabilities this summer. Google issued no security patches in its update last month.","feed":"cyberscoop","publisher":"cyberscoop","topics":["remediation"],"type":"analysis"},{"id":"f8bd4ffd740b","published":"2025-08-04T20:46:32.000Z","title":"GHSA-fm3m-jrgm-5ppg / CVE-2025-53534: RatPanel can perform remote command execution without authorization","url":"https://github.com/advisories/GHSA-fm3m-jrgm-5ppg","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"0eb579a4a6da","published":"2025-08-04T20:28:36.000Z","title":"GHSA-h5rc-j5f5-3gcm / CVE-2025-54804: russh is missing overflow checks during channel windows adjust","url":"https://github.com/advisories/GHSA-h5rc-j5f5-3gcm","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"901cdd9ee350","published":"2025-08-04T20:01:00.000Z","title":"Highlight & Summarize: RAG without the jailbreaks","url":"https://arxiv.org/abs/2508.02872","summary":"Preventing jailbreaking and model hijacking of Large Language Models (LLMs) is an important yet challenging task. When interacting with a chatbot, malicious users can input specially crafted prompts that cause the LLM to generate undesirable content or perform a different task…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"140a35522e1f","published":"2025-08-04T19:22:04.000Z","title":"Google says its AI-based bug hunter found 20 security vulnerabilities","url":"https://techcrunch.com/2025/08/04/google-says-its-ai-based-bug-hunter-found-20-security-vulnerabilities","summary":"The discoveries by an AI-based bug hunter are significant, as it shows these tools are starting to get real results, even if they still need a human.","feed":"techcrunch-security","publisher":"techcrunch-security","topics":[],"type":"analysis"},{"id":"647dca6c127f","published":"2025-08-04T19:09:00.000Z","title":"Proton fixes Authenticator bug leaking TOTP secrets in logs","url":"https://bleepingcomputer.com/news/security/proton-fixes-authenticator-bug-leaking-totp-secrets-in-logs","summary":"Proton fixed a bug in its new Authenticator app for iOS that logged users' sensitive TOTP secrets in plaintext, potentially exposing multi-factor authentication codes if the logs were shared.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"f95fcfbd4a86","published":"2025-08-04T18:44:50.000Z","title":"Automated Code Repair for C/C++ Static Analysis Alerts","url":"https://arxiv.org/abs/2508.02820","summary":"(Note: This work is a preprint.) Static analysis (SA) tools produce many diagnostic alerts indicating that source code in C or C++ may be defective and potentially vulnerable to security exploits. Many of these alerts are false positives. Identifying the true-positive alerts…","feed":"arxiv-remediation","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"b34854439728","published":"2025-08-04T18:27:26.000Z","title":"Large Reasoning Models Are Autonomous Jailbreak Agents","url":"https://arxiv.org/abs/2508.04039","summary":"Jailbreaking -- bypassing built-in safety mechanisms in AI models -- has traditionally required complex technical procedures or specialized human expertise. In this study, we show that the persuasive capabilities of large reasoning models (LRMs) simplify and scale jailbreaking,…","feed":"arxiv-agentic","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"0344c6ffcc39","published":"2025-08-04T16:40:31.000Z","title":"Quoting @himbodhisattva","url":"https://simonwillison.net/2025/Aug/4/himbodhisattva","summary":"for services that wrap GPT-3, is it possible to do the equivalent of sql injection? like, a prompt-injection attack? make it think it's completed the task and then get access to the generation, and ask it to repeat the original instruction? — @himbodhisattva , coining the term…","feed":"simonw","publisher":"simonw","topics":["ai-security"],"type":"analysis"},{"id":"54f022670316","published":"2025-08-04T16:07:05.000Z","title":"GHSA-65fc-cr5f-v7r2 / CVE-2025-54803: js-toml Prototype Pollution Vulnerability","url":"https://github.com/advisories/GHSA-65fc-cr5f-v7r2","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"e38c03c8002f","published":"2025-08-04T16:06:00.002Z","title":"NVIDIA Triton Bugs Let Unauthenticated Attackers Execute Code and Hijack AI Servers","url":"https://thehackernews.com/2025/08/nvidia-triton-bugs-let-unauthenticated.html","summary":"A newly disclosed set of security flaws in NVIDIA's Triton Inference Server for Windows and Linux, an open-source platform for running artificial intelligence (AI) models at scale, could be exploited to take over susceptible servers. \"When chained together, these flaws can…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"incident"},{"id":"ed4371e89438","published":"2025-08-04T15:57:48.000Z","title":"GHSA-48rp-jc79-2264 / CVE-2025-54802: pyLoad CNL Blueprint allows Path Traversal through `dlc_path` which leads to Remote Code Execution (RCE)","url":"https://github.com/advisories/GHSA-48rp-jc79-2264","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"17a3716647b4","published":"2025-08-04T15:22:23.000Z","title":"GHSA-5662-2rj7-f2v6 / CVE-2025-54796: copyparty allows Regex Denial of Service (ReDoS) in the upload listing","url":"https://github.com/advisories/GHSA-5662-2rj7-f2v6","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"f2efe1071659","published":"2025-08-04T15:18:03.000Z","title":"GHSA-x56v-x2h6-7j34 / CVE-2025-54795: Claude Code echo command allowed bypass of user approval prompt for command execution","url":"https://github.com/advisories/GHSA-x56v-x2h6-7j34","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca","agentic-sdlc"],"type":"advisory"},{"id":"7ee2b5e3e21e","published":"2025-08-04T15:15:52.000Z","title":"GHSA-pmw4-pwvc-3hx2 / CVE-2025-54794: Claude Code Research Preview has a Path Restriction Bypass which could allow unauthorized file access","url":"https://github.com/advisories/GHSA-pmw4-pwvc-3hx2","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca","agentic-sdlc"],"type":"advisory"},{"id":"36e63f1507a8","published":"2025-08-04T15:12:03.000Z","title":"GHSA-vf2r-cxg9-p7rf / CVE-2025-54119: The ADOdb sqlite3 driver allows SQL injection","url":"https://github.com/advisories/GHSA-vf2r-cxg9-p7rf","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"9972870463a5","published":"2025-08-04T14:58:39.000Z","title":"From Commits to Confidence: Towards Stability-Informed Risk Assessment in Open Source Software","url":"https://arxiv.org/abs/2508.02487","summary":"Open source software (OSS) generates trillions of dollars in economic value and has become essential to the technical infrastructures that power organizations worldwide. As these systems increasingly depend on OSS, understanding the evolution of these projects is critical.…","feed":"arxiv-sca","publisher":"arxiv","topics":["trust"],"type":"research"},{"id":"97a57af4f33d","published":"2025-08-04T14:48:25.000Z","title":"GHSA-mm3p-j368-7jcr / CVE-2025-54387: IPX Allows Path Traversal via Prefix Matching Bypass","url":"https://github.com/advisories/GHSA-mm3p-j368-7jcr","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"a226f5c94bea","published":"2025-08-04T12:55:59.000Z","title":"Breaking NVIDIA Triton: CVE-2025-23319 - A Vulnerability Chain Leading to AI Server Takeover","url":"https://wiz.io/blog/nvidia-triton-cve-2025-23319-vuln-chain-to-ai-server","summary":"Wiz Research discovers a critical vulnerability chain allowing unauthenticated attackers to take over NVIDIA's Triton Inference Server.","feed":"wiz","publisher":"wiz","topics":[],"type":"research"},{"id":"149b5d8f0bff","published":"2025-08-04T11:55:00.002Z","title":"⚡ Weekly Recap: VPN 0-Day, Encryption Backdoor, AI Malware, macOS Flaw, ATM Hack & More","url":"https://thehackernews.com/2025/08/weekly-recap-vpn-0-day-encryption.html","summary":"Malware isn’t just trying to hide anymore—it’s trying to belong. We’re seeing code that talks like us, logs like us, even documents itself like a helpful teammate. Some threats now look more like developer tools than exploits. Others borrow trust from open-source platforms, or…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"roundup"},{"id":"9c886003cb5a","published":"2025-08-04T11:28:34.000Z","title":"Data Security in Large Language Models: Risks, Defense, and Directions","url":"https://arxiv.org/abs/2508.02312","summary":"Large Language Models (LLMs), now a foundation in advancing natural language processing, power applications such as text generation, machine translation, and conversational systems. Despite their transformative potential, these models inherently rely on massive amounts of…","feed":"arxiv-agentic","publisher":"arxiv","topics":["models","ai-security"],"type":"research"},{"id":"53d8aa7e1756","published":"2025-08-04T11:26:00.000Z","title":"Ransomware gangs join attacks targeting Microsoft SharePoint servers","url":"https://bleepingcomputer.com/news/security/ransomware-gangs-join-attacks-targeting-microsoft-sharepoint-servers","summary":"Ransomware gangs have recently joined ongoing attacks targeting a Microsoft SharePoint vulnerability chain, part of a broader exploitation campaign that has already led to the breach of at least 148 organizations worldwide.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"78fed8c9e454","published":"2025-08-04T10:00:00.000Z","title":"Mozilla warns of phishing attacks targeting add-on developers","url":"https://bleepingcomputer.com/news/security/mozilla-warns-of-phishing-attacks-targeting-add-on-developers","summary":"Mozilla has warned browser extension developers of an active phishing campaign targeting accounts on its official AMO (addons.mozilla.org) repository.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"analysis"},{"id":"6ac79d844b4d","published":"2025-08-04T09:30:28.000Z","title":"GHSA-3c93-92r7-j934 / CVE-2025-8341: Grafana Infinity Datasource Plugin SSRF Vulnerability","url":"https://github.com/advisories/GHSA-3c93-92r7-j934","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"4979aa152a93","published":"2025-08-04T08:57:14.000Z","title":"Can LLMs effectively provide game-theoretic-based scenarios for cybersecurity?","url":"https://arxiv.org/abs/2508.05670","summary":"Game theory has long served as a foundational tool in cybersecurity to test, predict, and design strategic interactions between attackers and defenders. The recent advent of Large Language Models (LLMs) offers new tools and challenges for the security of computer systems; In…","feed":"arxiv-agentic","publisher":"arxiv","topics":[],"type":"research"},{"id":"2d811db488d3","published":"2025-08-04T08:31:56.000Z","title":"Whispering Agents: An Event-driven Covert Communication Protocol For the Internet of Agents","url":"https://arxiv.org/abs/2508.02188","summary":"The emergence of the Internet of Agents (IoA) introduces critical challenges for communication privacy in sensitive, high-stakes domains. While standard Agent-to-Agent (A2A) protocols secure message content, they are not designed to protect the act of communication itself,…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"aee349eef1ad","published":"2025-08-04T07:34:00.000Z","title":"FedLAD: A Linear Algebra Based Data Poisoning Defence for Federated Learning","url":"https://arxiv.org/abs/2508.02136","summary":"Sybil attacks pose a significant threat to federated learning, as malicious nodes can collaborate and gain a majority, thereby overwhelming the system. Therefore, it is essential to develop countermeasures that ensure the security of federated learning environments. We present…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"78683aeb7f93","published":"2025-08-04T07:13:18.000Z","title":"Towards Stealthy and Effective Backdoor Attacks on Lane Detection: A Naturalistic Data Poisoning Approach","url":"https://arxiv.org/abs/2508.15778","summary":"Deep learning-based lane detection (LD) plays a critical role in autonomous driving and advanced driver assistance systems. However, its vulnerability to backdoor attacks presents a significant security concern. Existing backdoor attack methods on LD often exhibit limited…","feed":"arxiv-sca","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"427b445be8c6","published":"2025-08-04T07:04:58.000Z","title":"Cursor IDE: Arbitrary Data Exfiltration Via Mermaid (CVE-2025-54132)","url":"https://embracethered.com/blog/posts/2025/cursor-data-exfiltration-with-mermaid","summary":"Cursor is a popular AI code editor. In this post I want to share how I found an interesting data exfiltration issue, the demo exploits built and how it got fixed. When using Cursor I noticed that it can render Mermaid diagrams. Cursor Renders Mermaid Diagrams If you are not…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["agentic-sdlc"],"type":"research"},{"id":"9737b4fbfc13","published":"2025-08-04T06:38:59.000Z","title":"Attractive Metadata Attack: Inducing LLM Agents to Invoke Malicious Tools","url":"https://arxiv.org/abs/2508.02110","summary":"Large language model (LLM) agents have demonstrated remarkable capabilities in complex reasoning and decision-making by leveraging external tools. However, this tool-centric paradigm introduces a previously underexplored attack surface, where adversaries can manipulate tool…","feed":"arxiv-aisec","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"21eecbe22409","published":"2025-08-04T06:30:16.000Z","title":"GHSA-mx88-9p24-6jmg: Malware in epic-fortnite-shared-values","url":"https://github.com/advisories/GHSA-mx88-9p24-6jmg","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-malware","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ce528ccbf2f3","published":"2025-08-04T06:29:22.000Z","title":"GHSA-mprg-wf28-hgqq: Malware in epic-games-nav-share","url":"https://github.com/advisories/GHSA-mprg-wf28-hgqq","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-malware","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"e854e4a852f0","published":"2025-08-04T06:00:22.000Z","title":"FPEdit: Robust LLM Fingerprinting through Localized Parameter Editing","url":"https://arxiv.org/abs/2508.02092","summary":"Large language models represent significant investments in computation, data, and engineering expertise, making them extraordinarily valuable intellectual assets. Nevertheless, these AI assets remain vulnerable to unauthorized redistribution and commercial exploitation through…","feed":"arxiv-models","publisher":"arxiv","topics":[],"type":"research"},{"id":"ce288ad12b92","published":"2025-08-04T02:27:14.000Z","title":"DIRF: A Framework for Digital Identity Protection and Clone Governance in Agentic AI Systems","url":"https://arxiv.org/abs/2508.01997","summary":"The rapid advancement and widespread adoption of generative artificial intelligence (AI) pose significant threats to the integrity of personal identity, including digital cloning, sophisticated impersonation, and the unauthorized monetization of identity-related data.…","feed":"arxiv-agentic","publisher":"arxiv","topics":["trust"],"type":"research"},{"id":"d1bb3a8b167c","published":"2025-08-03T13:16:18.000Z","title":"LLM-Assisted Model-Based Fuzzing of Protocol Implementations","url":"https://arxiv.org/abs/2508.01750","summary":"Testing network protocol implementations is critical for ensuring the reliability, security, and interoperability of distributed systems. Faults in protocol behavior can lead to vulnerabilities and system failures, especially in real-time and mission-critical applications. A…","feed":"arxiv-dast","publisher":"arxiv","topics":["dast","agentic-sdlc"],"type":"research"},{"id":"060493446ed5","published":"2025-08-03T12:30:30.000Z","title":"GHSA-xg8j-j6vp-6h5w / CVE-2024-51775: Apache Zeppelin: Missing Origin Validation in WebSockets vulnerability","url":"https://github.com/advisories/GHSA-xg8j-j6vp-6h5w","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"0eeebd56c566","published":"2025-08-03T12:30:30.000Z","title":"GHSA-p288-459w-jxj6 / CVE-2024-41177: Apache Zeppelin: XSS in the Helium module","url":"https://github.com/advisories/GHSA-p288-459w-jxj6","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"3563a3e4704b","published":"2025-08-03T12:30:30.000Z","title":"GHSA-jr43-q92q-5q82 / CVE-2024-52279: Apache Zeppelin: Arbitrary file read by adding malicious JDBC connection string","url":"https://github.com/advisories/GHSA-jr43-q92q-5q82","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"2fe3305a7938","published":"2025-08-03T08:30:58.000Z","title":"Anthropic Filesystem MCP Server: Directory Access Bypass via Improper Path Validation","url":"https://embracethered.com/blog/posts/2025/anthropic-filesystem-mcp-server-bypass","summary":"A few months ago I was looking at the filesystem MCP server from Anthropic. The server allows to give an AI, like Claude Desktop, access to the local filesystem to read files or edit them and so forth. I was curious about access control and in the documentation there is a…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":[],"type":"analysis"},{"id":"5970f631ae18","published":"2025-08-03T08:26:13.000Z","title":"JSidentify-V2: Leveraging Dynamic Memory Fingerprinting for Mini-Game Plagiarism Detection","url":"https://arxiv.org/abs/2508.01655","summary":"The explosive growth of mini-game platforms has led to widespread code plagiarism, where malicious users access popular games' source code and republish them with modifications. While existing static analysis tools can detect simple obfuscation techniques like variable renaming…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"790d174e5ac9","published":"2025-08-03T08:12:21.000Z","title":"DUP: Detection-guided Unlearning for Backdoor Purification in Language Models","url":"https://arxiv.org/abs/2508.01647","summary":"As backdoor attacks become more stealthy and robust, they reveal critical weaknesses in current defense strategies: detection methods often rely on coarse-grained feature statistics, and purification methods typically require full retraining or additional clean models. To…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"4357da076706","published":"2025-08-03T05:42:20.000Z","title":"Practical, Generalizable and Robust Backdoor Attacks on Text-to-Image Diffusion Models","url":"https://arxiv.org/abs/2508.01605","summary":"Text-to-image diffusion models (T2I DMs) have achieved remarkable success in generating high-quality and diverse images from text prompts, yet recent studies have revealed their vulnerability to backdoor attacks. Existing attack methods suffer from critical limitations: 1) they…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"38ca1dc13045","published":"2025-08-02T20:19:22.000Z","title":"Reconstructing Trust Embeddings from Siamese Trust Scores: A Direct-Sum Approach with Fixed-Point Semantics","url":"https://arxiv.org/abs/2508.01479","summary":"We study the inverse problem of reconstructing high-dimensional trust embeddings from the one-dimensional Siamese trust scores that many distributed-security frameworks expose. Starting from two independent agents that publish time-stamped similarity scores for the same set of…","feed":"arxiv-agentic","publisher":"arxiv","topics":[],"type":"research"},{"id":"4d23d8007678","published":"2025-08-02T17:57:46.000Z","title":"Think Broad, Act Narrow: CWE Identification with Multi-Agent Large Language Models","url":"https://arxiv.org/abs/2508.01451","summary":"Machine learning and Large language models (LLMs) for vulnerability detection has received significant attention in recent years. Unfortunately, state-of-the-art techniques show that LLMs are unsuccessful in even distinguishing the vulnerable function from its benign…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"9c04cc9102ef","published":"2025-08-02T13:52:15.000Z","title":"Prompt to Pwn: Automated Exploit Generation for Smart Contracts","url":"https://arxiv.org/abs/2508.01371","summary":"Smart contracts are important for digital finance, yet they are hard to patch once deployed. Prior work has mainly explored LLMs for smart contract vulnerability detection, leaving end-to-end automated exploit generation (AEG) much less understood. We study that gap with…","feed":"arxiv-crse","publisher":"arxiv","topics":[],"type":"research"},{"id":"eabedcab4c96","published":"2025-08-02T13:38:04.000Z","title":"ConfGuard: A Simple and Effective Backdoor Detection for Large Language Models","url":"https://arxiv.org/abs/2508.01365","summary":"Backdoor attacks pose a significant threat to Large Language Models (LLMs), where adversaries can embed hidden triggers to manipulate LLM's outputs. Most existing defense methods, primarily designed for classification tasks, are ineffective against the autoregressive nature and…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"1bd79fd585f5","published":"2025-08-02T13:14:10.000Z","title":"An Empirical Validation of Open Source Repository Stability Metrics","url":"https://arxiv.org/abs/2508.01358","summary":"Over the past few decades, open source software has been continuously integrated into software supply chains worldwide, drastically increasing reliance and dependence. Because of the role this software plays, it is important to understand ways to measure and promote its…","feed":"arxiv-sca","publisher":"arxiv","topics":["sca"],"type":"research"},{"id":"ab1d55670f41","published":"2025-08-02T12:56:27.000Z","title":"NATLM: Detecting Defects in NFT Smart Contracts Leveraging LLM","url":"https://arxiv.org/abs/2508.01351","summary":"Security issues are becoming increasingly significant with the rapid evolution of Non-fungible Tokens (NFTs). As NFTs are traded as digital assets, they have emerged as prime targets for cyber attackers. In the development of NFT smart contracts, there may exist undiscovered…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"0f0b73a2e4bd","published":"2025-08-02T12:49:41.000Z","title":"MultiCFV: Detecting Control Flow Vulnerabilities in Smart Contracts Leveraging Multimodal Deep Learning","url":"https://arxiv.org/abs/2508.01346","summary":"The introduction of smart contract functionality marks the advent of the blockchain 2.0 era, enabling blockchain technology to support digital currency transactions and complex distributed applications. However, many smart contracts have been found to contain vulnerabilities…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast"],"type":"research"},{"id":"af309d34892e","published":"2025-08-02T11:59:21.000Z","title":"BlockA2A: Towards Secure and Verifiable Agent-to-Agent Interoperability","url":"https://arxiv.org/abs/2508.01332","summary":"The rapid adoption of agentic AI, powered by large language models (LLMs), is transforming enterprise ecosystems with autonomous agents that execute complex workflows. Yet we observe several key security vulnerabilities in LLM-driven multi-agent systems (MASes): fragmented…","feed":"arxiv-agentic","publisher":"arxiv","topics":["trust","models"],"type":"research"},{"id":"76560c609d2a","published":"2025-08-02T10:36:01.000Z","title":"PUZZLED: Jailbreaking LLMs through Word-Based Puzzles","url":"https://arxiv.org/abs/2508.01306","summary":"As large language models (LLMs) are increasingly deployed across diverse domains, ensuring their safety has become a critical concern. In response, studies on jailbreak attacks have been actively growing. Existing approaches typically rely on iterative prompt engineering or…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"ac78cf60a00a","published":"2025-08-02T07:59:34.000Z","title":"AgentArmor: Enforcing Program Analysis on Agent Runtime Trace to Defend Against Prompt Injection","url":"https://arxiv.org/abs/2508.01249","summary":"Large Language Model (LLM) agents offer a powerful new paradigm for solving various problems by combining natural language reasoning with the execution of external tools. However, their dynamic and non-transparent behavior introduces critical security risks, particularly in the…","feed":"arxiv-sca","publisher":"arxiv","topics":["sast","ai-security"],"type":"research"},{"id":"f077a0514c5b","published":"2025-08-02T07:31:58.000Z","title":"Turning ChatGPT Codex Into A ZombAI Agent","url":"https://embracethered.com/blog/posts/2025/chatgpt-codex-remote-control-zombai","summary":"Today we cover ChatGPT Codex as part of the Month of AI Bugs series. ChatGPT Codex is a cloud-based software engineering agent that answers codebase questions, executes code, and drafts pull requests. In particular, this post will demonstrate how Codex is vulnerable to prompt…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["sca","agentic-sdlc","ai-security"],"type":"analysis"},{"id":"67bac9c8cb5b","published":"2025-08-02T06:56:00.002Z","title":"Akira Ransomware Exploits SonicWall VPNs in Likely Zero-Day Attack on Fully-Patched Devices","url":"https://thehackernews.com/2025/08/akira-ransomware-exploits-sonicwall.html","summary":"SonicWall SSL VPN devices have become the target of Akira ransomware attacks as part of a newfound surge in activity observed in late July 2025. \"In the intrusions reviewed, multiple pre-ransomware intrusions were observed within a short period of time, each involving VPN…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation"],"type":"incident"},{"id":"734d239cd85e","published":"2025-08-02T01:43:39.000Z","title":"AgentSight: System-Level Observability for AI Agents Using eBPF","url":"https://arxiv.org/abs/2508.02736","summary":"Modern software infrastructure increasingly relies on LLM agents for development and maintenance, such as Claude Code and Gemini-cli. However, these AI agents differ fundamentally from traditional deterministic software, posing a significant challenge to conventional monitoring…","feed":"arxiv-agentic","publisher":"arxiv","topics":["agentic-sdlc"],"type":"research"},{"id":"048dfbb06eb8","published":"2025-08-01T21:37:16.000Z","title":"Provably Secure Retrieval-Augmented Generation","url":"https://arxiv.org/abs/2508.01084","summary":"Although Retrieval-Augmented Generation (RAG) systems have been widely applied, the privacy and security risks they face, such as data leakage and data poisoning, have not been systematically addressed yet. Existing defense strategies primarily rely on heuristic filtering or…","feed":"arxiv-models","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"3732148b3007","published":"2025-08-01T20:34:36.000Z","title":"CP-FREEZER: Latency Attacks against Vehicular Cooperative Perception","url":"https://arxiv.org/abs/2508.01062","summary":"Cooperative perception (CP) enhances situational awareness of connected and autonomous vehicles by exchanging and combining messages from multiple agents. While prior work has explored adversarial integrity attacks that degrade perceptual accuracy, little is known about CP's…","feed":"arxiv-agentic","publisher":"arxiv","topics":["models"],"type":"research"},{"id":"70b61a0f4c1e","published":"2025-08-01T20:11:58.000Z","title":"Autonomous Penetration Testing: Solving Capture-the-Flag Challenges with LLMs","url":"https://arxiv.org/abs/2508.01054","summary":"This study evaluates the ability of GPT-4o to autonomously solve beginner-level offensive security tasks by connecting the model to OverTheWire's Bandit capture-the-flag game. Of the 25 levels that were technically compatible with a single-command SSH framework, GPT-4o solved…","feed":"arxiv-dast","publisher":"arxiv","topics":["dast"],"type":"research"},{"id":"b64cb35ec35f","published":"2025-08-01T19:51:02.000Z","title":"Cursor’s AI coding agent morphed ‘into local shell’ with one-line prompt attack","url":"https://cyberscoop.com/cursor-ai-prompt-injection-attack-remote-code-privileges-aimlabs","summary":"The flaw, disclosed a month after it was patched, provided an attacker with remote code execution privileges by poisoning the data ingested by the model.","feed":"cyberscoop","publisher":"cyberscoop","topics":["agentic-sdlc","remediation","ai-security"],"type":"analysis"},{"id":"4cadb26830c9","published":"2025-08-01T18:43:13.000Z","title":"GHSA-85cg-cmq5-qjm7 / CVE-2025-54782: @nestjs/devtools-integration: CSRF to Sandbox Escape Allows for RCE against JS Developers","url":"https://github.com/advisories/GHSA-85cg-cmq5-qjm7","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"1ce82e1ced29","published":"2025-08-01T18:31:19.000Z","title":"GHSA-mr4h-qf9j-f665 / CVE-2025-6000: Hashicorp Vault has Code Execution Vulnerability via Plugin Configuration","url":"https://github.com/advisories/GHSA-mr4h-qf9j-f665","summary":"GitHub Advisory Database, severity critical.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"3ac26c5c5f62","published":"2025-08-01T18:31:19.000Z","title":"GHSA-qv3p-fmv3-9hww / CVE-2025-6014: Hashicorp Vault's TOTP Secrets Engine Susceptible to Code Reuse ","url":"https://github.com/advisories/GHSA-qv3p-fmv3-9hww","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"9d796872cc27","published":"2025-08-01T18:31:19.000Z","title":"GHSA-mwgr-84fv-3jh9 / CVE-2025-6011: Hashicorp Vault has an Observable Discrepancy on Existing and Non-Existing Users","url":"https://github.com/advisories/GHSA-mwgr-84fv-3jh9","summary":"GitHub Advisory Database, severity low.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"bbd1cb1f97b6","published":"2025-08-01T18:31:19.000Z","title":"GHSA-6c5r-4wfc-3mcx / CVE-2025-6037: Hashicorp Vault has Incorrect Validation for Non-CA Certificates","url":"https://github.com/advisories/GHSA-6c5r-4wfc-3mcx","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"bbe0aa4d4776","published":"2025-08-01T18:31:19.000Z","title":"GHSA-v6r4-35f9-9rpw / CVE-2025-6015: Hashicorp Vault has Login MFA Rate Limit Bypass Vulnerability","url":"https://github.com/advisories/GHSA-v6r4-35f9-9rpw","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"c34fdf372484","published":"2025-08-01T18:31:19.000Z","title":"GHSA-6h4p-m86h-hhgh / CVE-2025-5999: Hashicorp Vault has Privilege Escalation Vulnerability","url":"https://github.com/advisories/GHSA-6h4p-m86h-hhgh","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"ddd264d6438b","published":"2025-08-01T18:31:19.000Z","title":"GHSA-qgj7-fmq2-6cc4 / CVE-2025-6004: Hashicorp Vault has Lockout Feature Authentication Bypass","url":"https://github.com/advisories/GHSA-qgj7-fmq2-6cc4","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"7ca0f1ff6eb5","published":"2025-08-01T18:31:18.000Z","title":"GHSA-mvj3-hc7j-vp74 / CVE-2025-51502: Microweber has Reflected XSS Vulnerability in the layout Parameter","url":"https://github.com/advisories/GHSA-mvj3-hc7j-vp74","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"7e87fcefeafb","published":"2025-08-01T18:31:18.000Z","title":"GHSA-8357-fjvx-xrm8 / CVE-2025-51501: Microweber has Reflected XSS Vulnerability in the id Parameter","url":"https://github.com/advisories/GHSA-8357-fjvx-xrm8","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"eeb8ed10f0c6","published":"2025-08-01T18:31:18.000Z","title":"GHSA-2x2j-3c2v-g3c2 / CVE-2025-51504: Microweber XSS Vulnerability in the homepage Endpoint ","url":"https://github.com/advisories/GHSA-2x2j-3c2v-g3c2","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"04fad4259ebd","published":"2025-08-01T18:15:03.000Z","title":"GHSA-2rjv-cv85-xhgm: OpenSearch unauthorized data access on fields protected by field level security if field is a member of an object","url":"https://github.com/advisories/GHSA-2rjv-cv85-xhgm","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"d4f03d0ed23e","published":"2025-08-01T18:15:00.000Z","title":"GHSA-rrmm-wq7q-h4v5: OpenSearch unauthorized data access on fields protected by field masking for fields of type ip, geo_point, geo_shape, xy_point, xy_shape","url":"https://github.com/advisories/GHSA-rrmm-wq7q-h4v5","summary":"GitHub Advisory Database, severity medium.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"6c9467619fe7","published":"2025-08-01T18:10:21.000Z","title":"GHSA-8j63-96wh-wh3j / CVE-2025-54424: 1Panel agent certificate verification bypass leading to arbitrary command execution","url":"https://github.com/advisories/GHSA-8j63-96wh-wh3j","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"0e9081dec22b","published":"2025-08-01T18:08:15.000Z","title":"GHSA-q6gg-9f92-r9wg / CVE-2025-54386: Traefik Client Plugin's Path Traversal Vulnerability Allows Arbitrary File Overwrite and Remote Code Execution","url":"https://github.com/advisories/GHSA-q6gg-9f92-r9wg","summary":"GitHub Advisory Database, severity high.","feed":"ghsa-reviewed","publisher":"github-advisories","topics":["sca"],"type":"advisory"},{"id":"225e150f348d","published":"2025-08-01T17:28:00.000Z","title":"SonicWall firewall devices hit in surge of Akira ransomware attacks","url":"https://bleepingcomputer.com/news/security/surge-of-akira-ransomware-attacks-hits-sonicwall-firewall-devices","summary":"SonicWall firewall devices have been increasingly targeted since late July in a surge of Akira ransomware attacks, potentially exploiting a previously unknown security vulnerability, according to cybersecurity company Arctic Wolf.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"ade7dee26fc0","published":"2025-08-01T16:13:00.000Z","title":"Pi-hole discloses data breach triggered by WordPress plugin flaw","url":"https://bleepingcomputer.com/news/security/pi-hole-discloses-data-breach-via-givewp-wordpress-plugin-flaw","summary":"Pi-hole, a popular network-level ad-blocker, has disclosed that donor names and email addresses were exposed through a security vulnerability in the GiveWP WordPress donation plugin.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":[],"type":"incident"},{"id":"e83e9c22fad5","published":"2025-08-01T15:31:00.004Z","title":"Cursor AI Code Editor Fixed Flaw Allowing Attackers to Run Commands via Prompt Injection","url":"https://thehackernews.com/2025/08/cursor-ai-code-editor-fixed-flaw.html","summary":"Cybersecurity researchers have disclosed a now-patched, high-severity security flaw in Cursor, a popular artificial intelligence (AI) code editor, that could result in remote code execution (RCE). The vulnerability, tracked as CVE-2025-54135 (CVSS score: 8.6), has been…","feed":"thehackernews","publisher":"thehackernews","topics":["remediation","ai-security"],"type":"research"},{"id":"2aca4d2f7f20","published":"2025-08-01T15:00:58.000Z","title":"Exfiltrating Your ChatGPT Chat History and Memories With Prompt Injection","url":"https://embracethered.com/blog/posts/2025/chatgpt-chat-history-data-exfiltration","summary":"In this post we demonstrate how a bypass in OpenAI’s “safe URL” rendering feature allows ChatGPT to send personal information to a third-party server. This can be exploited by an adversary via a prompt injection via untrusted data. If you process untrusted content, like…","feed":"embrace-the-red","publisher":"embrace-the-red","topics":["ai-security"],"type":"analysis"},{"id":"30657d9293c4","published":"2025-08-01T13:32:52.000Z","title":"🎉 Celebrating Five Years of OpenSSF: A Journey Through Open Source Security","url":"https://openssf.org/blog/2025/08/01/%f0%9f%8e%89-celebrating-five-years-of-openssf-a-journey-through-open-source-security","summary":"August 2025 marks five years since the official formation of the Open Source Security Foundation (OpenSSF). Born out of a critical need to secure the software supply chains and open...","feed":"openssf","publisher":"openssf","topics":["sca"],"type":"analysis"},{"id":"ba383c1f5917","published":"2025-08-01T13:21:15.000Z","title":"Authorities seize BlackSuit ransomware gang’s servers","url":"https://techcrunch.com/2025/08/01/authorities-seize-blacksuit-ransomware-gangs-servers","summary":"German authorities said they have seized the servers used by the long-running ransomware gang, BlackSuit, which is blamed for cyberattacks across Europe and the U.S., including the City of Dallas.","feed":"techcrunch-security","publisher":"techcrunch-security","topics":[],"type":"incident"},{"id":"22a33696c9b9","published":"2025-08-01T13:04:28.000Z","title":"LeakSealer: A Semisupervised Defense for LLMs Against Prompt Injection and Leakage Attacks","url":"https://arxiv.org/abs/2508.00602","summary":"The generalization capabilities of Large Language Models (LLMs) have led to their widespread deployment across various applications. However, this increased adoption has introduced several security threats, notably in the forms of jailbreaking and data leakage attacks.…","feed":"arxiv-crse","publisher":"arxiv","topics":["sast","models","ai-security"],"type":"research"},{"id":"1784d2951a6f","published":"2025-08-01T13:02:00.000Z","title":"Attackers Use Fake OAuth Apps with Tycoon Kit to Breach Microsoft 365 Accounts","url":"https://thehackernews.com/2025/08/attackers-use-fake-oauth-apps-with.html","summary":"Cybersecurity researchers have detailed a new cluster of activity where threat actors are impersonating enterprises with fake Microsoft OAuth applications to facilitate credential harvesting as part of account takeover attacks. \"The fake Microsoft 365 applications impersonate…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"incident"},{"id":"3684ea790fa3","published":"2025-08-01T13:00:00.000Z","title":"AI-powered Cursor IDE vulnerable to prompt-injection attacks","url":"https://bleepingcomputer.com/news/security/ai-powered-cursor-ide-vulnerable-to-prompt-injection-attacks","summary":"A vulnerability that researchers call CurXecute is present in almost all versions of the AI-powered code editor Cursor, and can be exploited to execute remote code with developer privileges.","feed":"bleepingcomputer","publisher":"bleepingcomputer","topics":["agentic-sdlc"],"type":"research"},{"id":"b3c8592336ec","published":"2025-08-01T12:20:00.004Z","title":"AI-Generated Malicious npm Package Drains Solana Funds from 1,500+ Before Takedown","url":"https://thehackernews.com/2025/08/ai-generated-malicious-npm-package.html","summary":"Cybersecurity researchers have flagged a malicious npm package that was generated using artificial intelligence (AI) and concealed a cryptocurrency wallet drainer. The package, @kodane/patch-manager, claims to offer \"advanced license validation and registry optimization…","feed":"thehackernews","publisher":"thehackernews","topics":["sca"],"type":"research"},{"id":"f55bb6283d5e","published":"2025-08-01T11:52:24.000Z","title":"Activation-Guided Local Editing for Jailbreaking Attacks","url":"https://arxiv.org/abs/2508.00555","summary":"Jailbreaking is an essential adversarial technique for red-teaming these models to uncover and patch security flaws. However, existing jailbreak methods face significant drawbacks. Token-level jailbreak attacks often produce incoherent or unreadable inputs and exhibit poor…","feed":"arxiv-models","publisher":"arxiv","topics":["remediation","models"],"type":"research"},{"id":"e80b8fdf8b56","published":"2025-08-01T11:00:00.000Z","title":"You Are What You Eat: Why Your AI Security Tools Are Only as Strong as the Data You Feed Them","url":"https://thehackernews.com/2025/08/you-are-what-you-eat-why-your-ai.html","summary":"Just as triathletes know that peak performance requires more than expensive gear, cybersecurity teams are discovering that AI success depends less on the tools they deploy and more on the data that powers them The junk food problem in cybersecurity Imagine a triathlete who…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"analysis"},{"id":"4c084162d292","published":"2025-08-01T09:53:06.000Z","title":"CyGATE: Game-Theoretic Cyber Attack-Defense Engine for Patch Strategy Optimization","url":"https://arxiv.org/abs/2508.00478","summary":"Modern cyber attacks unfold through multiple stages, requiring defenders to dynamically prioritize mitigations under uncertainty. While game-theoretic models capture attacker-defender interactions, existing approaches often rely on static assumptions and lack integration with…","feed":"arxiv-agentic","publisher":"arxiv","topics":["ai-security"],"type":"research"},{"id":"168aa5270447","published":"2025-08-01T08:44:00.004Z","title":"Storm-2603 Deploys DNS-Controlled Backdoor in Warlock and LockBit Ransomware Attacks","url":"https://thehackernews.com/2025/08/storm-2603-exploits-sharepoint-flaws-to.html","summary":"The threat actor linked to the exploitation of the recently disclosed security flaws in Microsoft SharePoint Server is using a bespoke command-and-control (C2) framework called AK47 C2 (also spelled ak47c2) in its operations. The framework includes at least two different types…","feed":"thehackernews","publisher":"thehackernews","topics":[],"type":"incident"},{"id":"26608cdd0bf3","published":"2025-08-01T00:00:00.000Z","title":"Malware-Resistant Python without the Guesswork","url":"https://chainguard.dev/unchained/malware-resistant-python-without-the-guesswork","summary":"The recent compromise of the num2words package never made it into Chainguard Libraries for Python. Get the breakdown from the team on our packages you can trust.","feed":"chainguard","publisher":"chainguard","topics":[],"type":"analysis"}]}