Anthropic launches 'Claude Security' for enterprises - the first major defensive product designed to keep up with AI-powered exploits that compress the time-to-attack to minutes

Anthropic launched Claude Security in public beta yesterday, an enterprise tool that scans code repositories for vulnerabilities, rates each finding's severity and confidence, and generates patch instructions that engineers can apply through Claude Code. The launch is direct response to Mythos and similar AI-driven offensive tools that have been compressing the time between vulnerability disclosure and active exploitation - LiteLLM was exploited 36 hours after disclosure last week, LMDeploy in 13 hours the week before. CrowdStrike, Microsoft Security, Palo Alto Networks, SentinelOne, Trend, and Wiz are integrating Claude Opus 4.7 into their platforms.

Check

If your organization holds a Claude Enterprise subscription, evaluate Claude Security against your existing static analysis tools this week.

Affected

Claude Enterprise customers can access Claude Security in public beta now via claude.ai/security or the Claude.ai sidebar. No API integration required. Team and Max access is coming soon. The deeper relevance is for any security team facing the new exploitation cadence: AI-driven offense has shrunk the patch window for several recent disclosures.

Fix

Pilot Claude Security on a non-critical repository first - point it at a side project before pointing it at production code. Scheduled scans give ongoing coverage rather than one-off audits. Pair the output with Claude Code on the Web to work through patches in a single session. For organizations not on Claude Enterprise: evaluate Aisle, Wiz Code, or GitHub Copilot Autofix on confidence rating and false positive rate.