Securing Python projects Supply Chain
summary
Supply-chain attacks rise ~700% year-over-year and PyPI reports roughly 40 malicious packages per day; the talk shows concrete defenses: adoption of the Salsa (SLSA-like) build levels to harden build platforms, use of TUF-style update protections (tough/TUF) for signed metadata, and sixstore signing that binds OIDC identities to artifacts with a Python client that works in CI. Pip audit (backed by OSV) finds installed vulnerabilities; examples include PyTorch dependency‑confusion incidents and successful attestations.
AI-written summary · about
chapters (17)
- 0:00Intro
- 1:43The real cost of a vulnerable supply chain
- 3:28SolarWinds attack
- 6:05Secure supply chain frameworks
- 8:10Software signing
- 12:18Vulnerability databases
- 13:33Vulnerabilities and PyPI
- 14:55PyPI and malicious packages
- 16:08SBOMS and VEX
- 18:32Python container images
- 19:25Scanning for vulnerabilities in source code
- 20:14Python community initiatives
- 22:23PEP 458 & PEP 480
- 25:14PEP 708: Extending the Repository API to Mitigate Dependency Confusion Attacks
- 26:10PEP 710: Recording the provenance of installed packages
- 28:12SLSA Supply-chain Levels for Software Artifacts
- 29:06Graph for Understanding Artifact Composition
related talks
10mPyPI Supply Chain Security
25m"Security considerations in Python Packaging"
18mMalicious Needle in a Haystack – PyPi Security Pitfalls
20mMalicious Needle in a Haystack - PyPi Security Pitfalls
25mBad actors vs our community: detecting software supply chain...
22mSofware Supply Chain Security Is Hard. But Does it Need to Be?
26mTesting Python Security
28mpip audit dozens of vulnerabilities after