Anthropic launches Claude Security public beta for enterprise code security
Anthropic just rolled out Claude Security, a new tool that scans company code for vulnerabilities and even suggests fixes.
Right now, it's in public beta for enterprise users of Claude, but wider access is coming soon.
This launch follows their earlier Project Glasswing, which focused on protecting open-source infrastructure.
Claude Security traces dataflow, rates severity
Claude Security acts like a digital security researcher: it scans entire codebases, traces how data moves, and spots weak points using Anthropic's Opus 4.7 model (which also blocks risky stuff like mass data leaks).
It gives each result a confidence rating and includes severity among the finding details so developers can focus on what matters most.
Plus, it fits easily into existing workflows with features like scheduled scans and smooth integrations.