That's the reason why open source projects should always implement a linter check in their code-review pipeline, which scans for unusual usage of invisible characters. This includes non-code files like documentation or test cases, because they can also be abused (as seen in the XZ toolchain attack).
Gesendet: Dienstag, 24. März 2026 um 06:39
Von: "Martin J. Dürst via Unicode" <[email protected]>
An: "Nitai Sasson" <[email protected]>, "Karl Williamson" <[email protected]>
CC: "[email protected]" <[email protected]>
Betreff: Re: Scientific American: Open Source infected with malware from invisible Unicode characters
GlassWorm attack already on October 18, 2025, but the Scientific
American article is from March 21, 2026. Apparently the original report
didn't get enough publicity.
Regards, Martin.
On 2026-03-23 07:37, Nitai Sasson via Unicode wrote:
> Thank you for sharing, this is quite interesting. I tried to find examples of how this actually works. Found this article: https://www.koi.ai/blog/glassworm-first-self-propagating-worm-using-invisible-code-hits-openvsx-marketplace
>
> The screenshot in it shows a clearly suspicious line of code: var decodedBytes = decode(' ... a very long invisible string ... ');
>
> So yes, the string is invisible, but it's not in itself executable. It needs to be decoded using a small amount of normal, visible and very suspicious code. So the claim that the vulnerability is invisible and can't be caught by normal code review seems a bit disingenuous. It's just a new way to obfuscate a string.
>
> I haven't found any other description of what compromised source code looks like in practice. So best I can tell, while this is really interesting, it's not as undetectable to the naked eye as it sounds.
>
> Still, very interesting! And if anyone has information that I haven't found, please share. Any technical dive into it would likely be a good read.
>
> - Nitai
>
> -------- Original Message --------
> On Sunday, 03/22/26 at 11:12 Karl Williamson via Unicode <[email protected]> wrote:
> Open-source software has an invisible vulnerability. Hackers have found it
> A cybercrime campaign called GlassWorm is hiding malware in invisible
> characters and spreading it through software that millions of developers
> rely on The danger in the code came from characters that are invisible
> to the human eye. In early March researchers at several security firms
> examined what looked like empty space and found hidden Unicode
> characters that decoded into a malicious program. Investigators soon
> traced hundreds of compromised open-source components spread across
> GitHub, npm and
>
> Read in Scientific American: https://apple.news/ACCjFPpifQlCNSMetYCJ2Dg
