State of Dependency Management 2025 | Endor Labs
- Published
- Record updated
Summary
Endor Labs' 2025 report examines how AI coding agents and Model Context Protocol (MCP) servers add software supply chain risk. Researchers analyzed 10,663 GitHub repositories implementing MCP servers and tested AI-generated dependency recommendations across PyPI, npm, Maven, and NuGet. They found that 49 % of imported dependency versions have known vulnerabilities and 34 % are hallucinated, meaning they do not exist in package registries.
Mitigation
Endor Labs recommends treating AI-generated code as untrusted third-party input and applying the same code review, SAST/SCA scanning, and dependency governance used for human development. Organizations should train developers in secure, spec-driven prompting, integrate security tools into AI workflows via MCP, and vet MCP servers using allowlists and continuous monitoring for typosquatting or brand-jacking.