Researchers Uncover 30+ Flaws in AI Coding Tools Enabling Data Theft and RCE Attacks

Security researchers discovered over 30 vulnerabilities in AI-powered Integrated Development Environments (IDEs) collectively named IDEsaster. These flaws combine prompt injection primitives with legitimate features to enable data exfiltration and remote code execution attacks. The vulnerabilities affect popular AI coding tools and represent a significant security risk for developers using these platforms. The research was conducted by security researcher Ari Marzouk (MaccariTA) and highlights emerging threats in AI-assisted development environments.