Attacker Hijacks AI Coding Assistant Session, Spreads Shai-Hulud Across About 100 Repositories
Summary: Mandiant says an attacker hijacked an active AI coding-assistant session at an unnamed SaaS provider, after the assistant recommended attacker-poisoned software that was accepted, and then spread the Shai-Hulud worm across about 100 internal code repositories. The attacker installed an infostealer through a poisoned PyPI package, stole GitHub OAuth tokens and repository secrets and source code, and poisoned a package in the company's official namespace that a second employee pulled, causing a second infection.
What's compelling: The entry point was a recommendation the developer accepted, not a phishing link or a zero-day, and the result was roughly 100 repositories compromised plus stolen OAuth tokens and product source code. Mandiant's case study does not say how the session was taken over or when, so the initial access path remains undocumented.
Read the original at The Hacker News ↗
- Topics
- Malware
- Related
- ARTEX and beyond: AI-enabled penetration testing platforms Cyber Risks · Nov 13, 2026Fake GTA 6 download goes after Steam logins and crypto wallet data Cyber Risks · Nov 13, 2026Operation Endgame Takes Down Gootloader, CastleRAT and StealC Servers, Seizes 6.4M Stolen Logins Cyber Risks · Nov 13, 2026