Vercel Open-Sources deepsec Framework with 1,000+ Sandbox Concurrency for Local AI Security Scanning

According to Beating, Vercel has open-sourced deepsec, an AI-driven security testing framework that enables developers to scan large codebases locally without exposing source code to external cloud services. The framework employs a multi-stage verification workflow: after initial regex filtering, an AI Agent traces data flows and generates reports; a second Agent then performs validation to eliminate false positives, keeping error rates between 10% and 20%. For large repositories requiring days of single-machine processing, deepsec distributes scanning tasks across Vercel Sandboxes, supporting concurrent execution of over 1,000 sandboxes. The system also offers plugin mechanisms allowing Agents to write custom regex matchers for project-specific authentication logic and data layers.
Disclaimer: The information on this page may come from third-party sources and is for reference only. It does not represent the views or opinions of Gate and does not constitute any financial, investment, or legal advice. Virtual asset trading involves high risk. Please do not rely solely on the information on this page when making decisions. For details, see the Disclaimer.
Comment
0/400
No comments