Invisible malicious code attacks 151 GitHub repos and VS Code — Glassworm attack uses blockchain to steal tokens, credentials, and secrets

by Chief Editor

The Rise of Invisible Threats: How Unicode Malware is Changing the Cybersecurity Landscape

A new wave of sophisticated cyberattacks is targeting developers and software supply chains, utilizing a deceptive technique: hiding malicious code within invisible Unicode characters. Recent reports from Aikido Security reveal that over 151 GitHub repositories were compromised between March 3 and March 9, 2026, by a threat actor known as Glassworm. This isn’t a new tactic – Aikido has been tracking this activity for nearly a year – but the scale and sophistication of the current campaign are raising serious concerns.

How Does Invisible Unicode Malware Operate?

The core of this attack lies in exploiting Unicode’s Private Use Area (PUA) characters. These characters, specifically ranges 0xFE00 through 0xFE0F and 0xE0100 through 0xE01EF, render as zero-width whitespace in most code editors and terminals. To the human eye, they appear as blank space. However, a cleverly crafted decoder can extract hidden bytes from these characters and execute them, effectively running malicious code. This makes detection incredibly difficult, as standard code reviews often miss these subtle injections.

The attackers are increasingly blending these malicious injections with legitimate code changes, such as version bumps and small refactors. Aikido Security suspects the use of large language models to generate these changes, making the malicious code even harder to spot. This represents a significant escalation in the attacker’s tactics.

Targets and Impact: From GitHub to npm and VS Code

The Glassworm campaign has expanded beyond GitHub, now affecting npm packages and VS Code extensions. Notable repositories compromised include those belonging to Wasmer, Reworm and anomalyco (the organization behind OpenCode and SST). The malware has previously been linked to the Solana blockchain, using it as a command-and-control channel to steal tokens, credentials, and sensitive information.

The potential impact is substantial. Compromised packages can introduce vulnerabilities into countless projects that rely on them, creating a ripple effect of security risks. The difficulty in detecting these attacks means that many affected repositories may remain compromised for extended periods.

A History of Invisible Attacks

Aikido Security first discovered malicious npm packages using this technique in March 2025. By October 2025, the attackers had moved to Open VSX and GitHub repositories. This ongoing campaign demonstrates the persistence and adaptability of the Glassworm threat actor.

What Can Developers Do to Protect Themselves?

Protecting against these invisible threats requires a multi-layered approach. Traditional code review is no longer sufficient. Here are some key recommendations:

  • Automated Scanning Tools: Utilize tools specifically designed to scan for invisible Unicode characters.
  • Dependency Scrutiny: Carefully examine package names and dependencies before incorporating them into projects.
  • Stay Updated: Preserve your development tools and dependencies up to date with the latest security patches.
  • Secure Development Practices: Implement robust security practices throughout the software development lifecycle.

Aikido Security provides tools like Safe Chain to help developers protect against malicious code installed via package managers. Their research highlights the importance of proactive security measures in the face of increasingly sophisticated threats.

The Future of Supply Chain Attacks

The Glassworm campaign signals a worrying trend: the increasing sophistication of supply chain attacks. Attackers are moving beyond traditional methods and leveraging subtle, hard-to-detect techniques. The use of large language models to generate convincing code changes further complicates the defense.

We can expect to witness more attacks that exploit obscure features of programming languages and development tools. The focus will likely shift towards techniques that bypass traditional security measures and blend seamlessly into legitimate codebases. This will require a fundamental shift in how developers and organizations approach software security.

FAQ

Q: What is Unicode and why is it being exploited?
A: Unicode is a character encoding standard that allows for the representation of a wide range of characters. Attackers are exploiting the Private Use Area within Unicode to hide malicious code in characters that are invisible to the human eye.

Q: Is my project at risk?
A: If you use packages from GitHub, npm, or VS Code extensions, there is a potential risk. Regularly scan your dependencies and implement the security measures outlined above.

Q: What is Glassworm?
A: Glassworm is the name given to the threat actor responsible for these attacks. They have been active since at least March 2025 and are known for their use of invisible Unicode characters.

Q: Can I rely on code reviews to detect these attacks?
A: Traditional code reviews are often ineffective against these attacks, as the malicious code is hidden within invisible characters.

Did you know? The Solana blockchain was previously used by Glassworm as a command-and-control channel, making takedown efforts more challenging.

Pro Tip: Regularly update your dependencies and use automated scanning tools to identify potential vulnerabilities in your projects.

Stay informed about the latest cybersecurity threats and best practices. Explore additional resources from Aikido Security and other leading security firms.

You may also like

Leave a Comment