
Malware leveraging public infrastructure like GitHub on the rise
ReversingLabs researchers have uncovered two novel techniques running on GitHub — one abusing GitHub Gists, another issuing commands through git commit messages.

ReversingLabs researchers have uncovered two novel techniques running on GitHub — one abusing GitHub Gists, another issuing commands through git commit messages.

More than 1,500 tokens were exposed, leaving millions of AI models and datasets vulnerable. Here's what your security team can learn from the compromise.

U.K. and U.S. cyber-watchdogs say to start with Secure by Design — but don't stop there. Here are key takeaways from their new guidelines.

SolarWinds, Codecov, and Kaseya showed how adversaries with access to a development environment can wreak havoc. Multifactor authentication is key, but it's not an "end-all solution."

Mature organizations recognize that their AppSec approach has to keep pace with modern development teams. Here's why.

While AI is mostly seen as opening a new front in the threat landscape, it will also be tapped to fight back with advanced threat hunting and more.

The Threat Modeling Manifesto explains the how. Now teams must map out an actionable threat modeling program. Here are 10 key tips to get started.

While the new White House EO is largely focused on foundational AI, security teams reviewing AI initiatives are still in the hot seat.

Streamlining Software Supply Chain Security: ReversingLabs Now Accessible via AWS Marketplace

Ingrained development patterns and legacy testing tools are holdovers from a reactive era of AppSec. Here's how how to move software security forward.

Rather than wasting cycles on non-exploitable or remediated security holes, teams should focus on exploitability, and look for compromises including malware and tampering. Here's why.

ReversingLabs researchers have discovered npm packages that hide scripts broadcasting messages of peace related to the conflicts in Ukraine and in Israel and the Gaza Strip.

Zero trust can benefit threat modeling, so why not extend it to your AppSec? Understand the key benefits and challenges.

Don't neutralize CI/CD business gains by failing to account for risk. Here are best practices to ensure that your software development pipeline is secure.

SLSA and Sigstore are a good first step toward protecting ML models from attack. But they're not a panacea.
Get your 14-day free trial of Spectra Assure
Get Free TrialMore about Spectra Assure Free Trial