What Changes When Your Software Supply Chain Includes AI Writing Your Code?
Summary: The article argues that in the roughly 20 months since the Model Context Protocol launched, AI tools, models, and agents have become load-bearing parts of software builds, so prompts are now a real input to the build and a real way to compromise it. It cites Shai-Hulud, a self-propagating malicious package campaign that spread through developer toolchains this year, as evidence that knowing what is in your code is necessary but no longer sufficient.
What's compelling: An AI coding assistant can suggest a dependency that a developer accepts without the package ever crossing a human's threat model, and an autonomous agent can reach for one MCP tool that reaches for another - so the provenance question now applies to the model, the agent, and the tooling, not only the artifact.
Read the original at The Hacker News ↗
- Topics
- Supply Chain & AppSec
- Related
- Kitesurf is now open source Cyber Risks · Nov 13, 2026Trained to betray: How post-training supply chain attacks hijack coding agents Cyber Risks · Nov 12, 2026ZDI-26-913: (0Day) Meta fairseq load_audio_path Code Injection Remote Code Execution Vulnerability Cyber Risks · Nov 11, 2026