ACCORDING to The Hacker News, Anthropic has begun to roll out Claude Code Security, a new security feature for Claude Code that can scan a user's software codebase for vulnerabilities and suggest patches. The capability is currently available in a limited research preview to Enterprise and Team customers.
Anthropic says Claude Code Security scans codebases for security vulnerabilities and suggests targeted software patches for human review, helping teams find and fix security issues that traditional methods might miss. It is described as countering attacks where threat actors weaponise the same tools to automate vulnerability discovery, by giving defenders an advantage and raising the security baseline.
The system goes beyond static analysis, reasoning about the codebase and data flows, with a multi-stage verification process to filter out false positives and a severity rating for prioritisation. Results are shown in a Claude Code Security dashboard, with a human-in-the-loop approach and a confidence rating for each finding, and nothing is applied without human approval.