
Read our full analysis at: research.jfrog.com/post/agen…

The JFrog Security Research Team empowers developers and companies to excel by identifying, prioritizing, and mitigating software risks.

Read our full analysis at: research.jfrog.com/post/agen…

🚨 Microsoft's PyPI package agent-framework was compromised. Versions 1.26.1 and 1.26.2 install a .pth hook that steals Azure, Azure DevOps, and GitHub Actions secrets on every Python start. Pin agent-framework 1.26.0 and rotate secrets if you installed either version.

Xray has been updated to reflect that OpenJDK 21, 25, 27 are also affected. Fixed versions are now out! 17.0.21.1, 21.0.13.1, 25.0.5.1 and 27.0.1.1

We'll keep updating as details emerge 👇 research.jfrog.com/post/java…

🚨 Java4Shell reported unauthenticated RCE (CVE-2026-114938) A vulnerability in OpenJDK 17.0.6+, rated CVSS 4.0 9.2 (Critical). No fixed version yet. Tech details due Nov 17, 1PM PT after Oracle's patch release. Inventory your Java runtimes now. JFrog Security Research is tracking this CVE, and it's already detected in JFrog Xray and JFrog Catalog.

⚠️ We’re noticing a concerning rise in GHSA false positives, and here’s the latest example disrupting the software supply chain. The popular npm package (@)univerjs-pro/sheets-pivot-chart (125K weekly downloads) had v1.5.0 and 1.5.1 incorrectly flagged as malicious and removed. The JFrog Security team independently analyzed the packages and confirmed they are safe - likely just an automated scanner choking on obfuscated code. False alarms like this cause unnecessary friction and panic for developers. We've reached out to GHSA to get the advisories corrected. Advisories: GHSA-h6jf-g8cj-x4v3 MAL-2026-19944 (OSV) #AppSec #SupplyChainSecurity #Malware #FalsePositive #univerjs

Read the full research and technical details: research.jfrog.com/post/accu…

🚨 Ongoing fake interview campaign uncovered. Applied for a job lately? Be careful with that take-home test. A single laptop compromised during a job interview reportedly kicked off the $6.3B Binance hack. JFrog Security Research just tracked a live fake-interview scam from start to finish. Scammers built an entire fake company just to trick engineers into running a malicious take-home test.

🚨 One HTTP request is all it takes to hijack an AI robot. JFrog Security Research found 5 CVEs in Reachy Mini, enabling unauthenticated RCE, root escalation, and physical takeover over local Wi-Fi or BLE. 🤖 Full technical analysis + PoC video 👇 jfrog.com/blog/reachy-http-r…


One skipped byte -> account takeover. 🚨 Vert.x skips validating the letter 'L', treating `Content-Xength` as a valid `Content-Length` (CVE-2026-102479). JFrog Research weaponized this "byte debt" with a dangling-byte primitive to bypass reverse proxy socket validation, demonstrating end-to-end response poisoning and session fixation behind real reverse proxies. Full exploit mechanics and affected open-source projects in our writeup 👇

🚨 testmetest goes beyond npm. JFrog Security Research identified 42 armed GitHub repositories across 11 legitimate owners hiding an npm publisher, plus a new native addon that decrypts its payload at runtime and stays dormant in CI, sandboxes, and analysis tools. The campaign spans 25 npm package names. Check whether you installed any testmetest or @iobandamwer packages and remove them if found. Full analysis: research.jfrog.com/post/beyo…

🔎 Read the full research and technical details: research.jfrog.com/post/brot… 🛠️ Get the open-source Brotli2Shell scanner: github.com/jfrog/vulnerabili…

🚨 Brotli2Shell goes beyond HTTP2Shell. CVE-2026-111497 is a CVSS 9.2 out-of-bounds write in Google’s Brotli decoder that has been exploited for RCE and impacts far beyond HTTP. JFrog reproduced RIP control in FFmpeg via a crafted JPEG XL file and identified additional attack paths in ClickHouse, .NET, and Apache Arrow, where some fixes were still unavailable at the time of writing. We’ve also released an open-source scanner to uncover potential exposure beyond vulnerable package versions. Full research and technical details in our blog 👇

We are actively monitoring Node2Shell / CVE-2026-111497. Xray customers can easily check if they are impacted through Xray Threat Hub or Xray Impact Search. JFrog Advanced Security customers can further check impact through Contextual Analysis

🔎 We break down the LiteLLM hijacking, the attack details, and what you need to know in our blog 👇 research.jfrog.com/post/lite…