Support for Python Packages Added to GitHub Security Alerts
18.7.18 securityweek  Security

GitHub announced on Thursday that developers will be warned if the Python packages used by their applications are affected by known vulnerabilities.

The code hosting service last year introduced a new feature, the Dependency Graph, that lists the libraries used by a project. It later extended it with a capability designed to alert developers when one of the software libraries used by their project has a known security hole.

The Dependency Graph and security alerts initially worked only for Ruby and JavaScript packages, but, as promised when the features were launched, GitHub has now also added support for Python packages.

“We’ve chosen to launch the new platform offering with a few recent vulnerabilities,” GitHub said in a blog post. “Over the coming weeks, we will be adding more historical Python vulnerabilities to our database.”

The security alerts feature is powered by information collected from the National Vulnerability Database (NVD) and other sources. When a new flaw is disclosed, GitHub identifies all repositories that use the affected version and informs their owners.

The security alerts are enabled by default for public repositories, but the owners of private repositories will have to manually enable the feature.

When a vulnerable library is detected, a “Known security vulnerability” alert will be displayed next to it in the Dependency Graph. Administrators can also configure email alerts, web notifications, and warnings via the user interface, and they can configure who should see the alerts.

GitHub reported in March that the introduction of the security alerts led to a significant decrease in the number of vulnerable libraries on the platform.

When the feature was launched, GitHub’s initial scan revealed over 4 million vulnerabilities across more than 500,000 repositories. Roughly two weeks after the first notifications were sent out, over 450,000 of the flaws were addressed by updating the impacted library or removing it altogether.