In the past few years, I’ve worked closely with enterprise security teams to improve their open source governance processes. One recurring theme I keep seeing is this: most organizations know they have issues with OSS component vulnerabilities—but they’re stuck when it comes to actually governing them.
To better understand this, we analyzed the top 20 most vulnerable open source components commonly found in enterprise Java stacks (e.g., jackson-databind
, shiro
, mysql-connector-java
) and realized something important:
Vulnerabilities aren’t just about CVE counts—they’re indicators of systemic governance blind spots.
Here’s the full article with breakdowns:
[From the Top 20 Open Source Component Vulnerabilities: Rethinking the Challenges of Open Source Security Governance](#)
In this post, I break down how the BadUSB attack works—starting from its origin at Black Hat 2014 to a hands-on implementation using an Arduino UNO and custom HID firmware. The attack exploits the USB protocol's lack of strict device type enforcement, allowing a USB stick to masquerade as a keyboard and inject malicious commands without user interaction.
The write-up covers:
If you're interested in hardware-based attack vectors, HID spoofing, or defending against stealthy USB threats, this deep-dive might be useful.
Demo video: https://youtu.be/xE9liN19m7o?si=OMcjSC1xjqs-53Vd
We recently completed an in-depth survey and analysis of the domestic software security market in China (2025 edition).
The report explores:
Although the data focuses on China, many of the findings resonate globally, especially regarding DevSecOps adoption and evolving security expectations.
If you're a security vendor, CISO, security engineer, or just interested in how software security needs are shifting in 2025, feel free to check it out.
Would love to hear your thoughts!