Malicious content in issues or pull requests can trick AI agents in CI/CD workflows into running privileged commands in an attack researchers nicknamed PromptPwnd. AI agents embedded in CI/CD ...
The proof of concept shows it's possible to upload malicious PyTorch releases to GitHub by exploiting insecure misconfigurations in GitHub Actions. A pair of security researchers managed to infiltrate ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results