Anthropic Launches Claude Code Security to Scan Codebases for Security Vulnerabilities

A new feature inside Claude Code enables developers and security teams to identify and remediate vulnerabilities across their codebases, known as Claude Code Security.

Currently available in a limited research preview, the tool offers AI-powered code scanning that goes beyond conventional static analysis by understanding how code actually works, not just what it contains.

Traditional tools often rely on rule-based scanning to check for known vulnerability patterns, such as hardcoded passwords or missing encryption.

While effective to a point, they often miss complex or context-dependent flaws, such as logic errors or insufficient access controls, that attackers can exploit.

Smarter Detection and Verification

Each vulnerability identified by Claude undergoes a multi-layered verification process. The system re-examines its own detections to minimize false positives and assigns severity and confidence ratings to each result.

Findings then appear in the Claude Code Security dashboard, where teams can review suggested patches and approve the final fixes.

Nothing is applied automatically; users maintain full control over remediation steps. Anthropic’s move follows a growing need for automation in vulnerability management.

Security teams worldwide face overwhelming backlogs as the number of identified software flaws continues to soar. The launch builds on more than a year of research by Anthropic into Claude’s cybersecurity potential.

By supplementing human expertise with AI-driven insights, Claude Code Security aims to ease that workload while maintaining accuracy and accountability.

Feature	Description
AI vulnerability detection	Scans code for security issues and suggests fixes for review.
Context-aware analysis	Understands data flow and component interactions to find complex flaws.
Multi-stage verification	Rechecks findings to reduce false positives.
Severity ratings	Assigns priority levels and confidence scores to issues.
Human review	Developers review and approve suggested fixes.
Security dashboard	Shows validated issues and tracks remediation progress.
Powered by Claude Opus 4.6	Uses the latest Claude model to detect serious vulnerabilities.
Defender-focused design	Helps security teams manage growing vulnerability backlogs.

Anthropic’s internal Frontier Red Team has tested these capabilities through hackathons, collaboration with the Pacific Northwest National Laboratory, and real-world applications.

Using Claude Opus 4.6, released earlier this year, the AI reportedly helped uncover over 500 previously undetected vulnerabilities in open-source projects, some of which had remained hidden for decades despite extensive human oversight.

As AI continues to transform both offensive and defensive cybersecurity operations, Anthropic positions Claude Code Security as a tool for defenders to stay ahead of emerging AI-enabled threats.

The company envisions a future in which most global codebases are continuously scanned by intelligent systems that predict and prevent attacks before they occur.

Enterprise and Team customers can now request early access to Claude Code Security. At the same time, open-source maintainers are invited to join a fast-tracked preview program.

Follow us on Google News, LinkedIn, and X for daily cybersecurity updates. Contact us to feature your stories.