Anthropic Launches Claude Code Security: AI Finds 500+ Vulnerabilities in Open Source Code

AI-Powered Security Scanning Arrives

Anthropic launched Claude Code Security on February 20, 2026, as a limited research preview for Enterprise and Team customers. The tool scans codebases for security vulnerabilities and suggests targeted software patches for human review.

How It Works

Claude Code Security reads and reasons about code the way a human security researcher would — understanding how components interact, tracing how data moves through applications, and catching complex vulnerabilities that rule-based tools miss. Unlike traditional static analysis, it understands the semantic intent behind code patterns.

500+ Vulnerabilities Discovered

The headline achievement: Claude Opus 4.6 found over 500 vulnerabilities in production open-source codebases — bugs that had gone undetected for decades, despite years of expert review. These aren't trivial issues; they represent deep logic flaws that conventional scanners consistently miss.

Human-In-The-Loop

Anthropic emphasizes safety: "Nothing is applied without human approval. Claude Code Security identifies problems and suggests solutions, but developers always make the call."

Enterprise Availability

The capability is available as a limited research preview to:

• Claude Enterprise customers
• Claude Team customers

Each scan provides detailed vulnerability reports with suggested patches, severity classifications, and exploitability assessments.

This launch positions Anthropic directly in the growing AI-assisted security market, competing with tools like GitHub's Copilot Security and Snyk's DeepCode AI.

Source: Anthropic | The Hacker News | VentureBeat | CyberScoop