Making frontier cybersecurity capabilities available to defenders

Anthropic launched Claude Code Security inside Claude Code on the web to scan codebases for security flaws.

The AI tool looks beyond rule-based static analysis, ranks issues by severity and confidence, and suggests fixes for human approval.

The preview is limited to Enterprise and Team customers, with faster access for open-source maintainers.

Anthropic says it is meant to help defenders keep up with vulnerability backlogs and AI-enabled attacks.