Researchers have identified 7,600 malicious GitHub repositories participating in the FakeGit campaign, with over 800 impersonating AI tools or Model Context Protocol servers to distribute SmartLoader malware.

The threat actors behind FakeGit employ social engineering tactics at scale. They clone legitimate open-source projects, create lookalike developer profiles, and craft convincing README files to establish trust. Victims who download what they believe are legitimate AI frameworks or MCP servers instead receive SmartLoader, a malware payload designed for arbitrary code execution and system compromise.

The campaign targets developers and organizations seeking to integrate AI capabilities or protocol implementations into their workflows. The sheer volume of repositories, 7,600 across GitHub's platform, demonstrates both the attacker's resources and the challenge GitHub faces in moderating malicious content at speed. The prevalence of AI-themed impersonation reflects developer interest in generative AI tools and emerging protocols like MCP, which Anthropic released for Claude integration.

SmartLoader functions as a downloader and code execution framework. Once installed, it grants attackers the ability to execute arbitrary commands on compromised systems, potentially leading to data theft, lateral movement within networks, or deployment of secondary payloads like ransomware or information stealers.

Organizations should audit their dependencies immediately. Developers should verify repository authenticity by checking contributor history, organization verification badges, and comparing URLs against official documentation. GitHub repositories with freshly created accounts, minimal commit history, or suspicious README formatting warrant extra scrutiny.

The discovery underscores a persistent supply-chain attack vector. Attackers continue to exploit developer workflows because repositories remain a trusted source for code dependencies. The FakeGit campaign scales this attack by automating repository creation and leveraging trending technologies like AI and MCP to increase click-through rates.

GitHub has removed the confirmed malicious repositories. However, new variants