🟠 High | Source: The Hacker News
Attackers have seeded nearly 7,600 malicious GitHub repositories — over 800 of which impersonate AI tools or Model Context Protocol (MCP) servers — to distribute a malware loader called SmartLoader. The campaign, dubbed FakeGit, uses cloned legitimate projects, convincing READMEs, and fake developer profiles to lure developers into downloading malicious ZIP archives. The scale and focus on AI-themed lures makes this a significant supply chain risk for development teams actively adopting AI tooling.
Security Architect’s Take: Audit your organisation’s developer tooling policies to ensure engineers are vetting GitHub repositories before use — specifically checking commit history, account age, star authenticity, and source provenance before integrating any AI or MCP-related packages. Consider enforcing dependency scanning and allowlisting in CI/CD pipelines to catch unsigned or unvetted third-party code.
Original advisory: FakeGit Campaign Uses 7,600 GitHub Repositories to Spread SmartLoader Malware