"The attacks, noticed by several vendors using automated security tools, happened on May 11, spreading rapidly through package ecosystems thanks to the worm capabilities of the automated Mini Shai-Hulud malware platform, analysis found."
"Instead of stealing maintainer credentials directly, the attackers exploited a risky trigger, pull_request_target. This allows third-party workflows to run automatically - a way of avoiding maintainer approval fatigue - but means that the maintainer's short-lived OIDC tokens become vulnerable to scraping."
"A striking feature of the attacks is the ease with which the threat group blamed for the attack, TeamPCP, was able to hijack the project's legitimate release pipelines by exploiting a mixture of maintainer misconfigurations and GitHub Actions weaknesses."
"The exact number of package versions caught up in the attack varies depending on the source; according to Aikido Security it was 373 across 169 package namespaces, while SafeDep said the number was 404 package versions across 170 npm packages, with two affecting PyPI."
Attacks detected by multiple vendors on May 11 spread quickly through package ecosystems using worm capabilities from the automated Mini Shai-Hulud malware platform. Reported impact varied by source, with Aikido Security identifying 373 package versions across 169 namespaces and SafeDep identifying 404 package versions across 170 npm packages, including two affecting PyPI. The threat group TeamPCP leveraged legitimate release pipeline hijacking rather than direct credential theft. Maintainer misconfigurations combined with GitHub Actions weaknesses enabled the compromise. The attackers exploited the pull_request_target trigger, which runs third-party workflows automatically and can expose short-lived OIDC tokens to scraping.
Read at InfoWorld
Unable to calculate read time
Collection
[
|
...
]