Claude Security has officially launched its public beta, providing Claude Enterprise customers with advanced tools to scan for code vulnerabilities and generate fixes using Opus 4.7. This initiative aims to enhance cybersecurity measures across various organizations.
Why it matters: The rapid evolution of AI in cybersecurity has made it crucial for organizations to bolster their defenses against potential threats. With models like Claude Opus 4.7, teams can now identify and address software vulnerabilities more effectively.
Claude Security, previously known as Claude Code Security, has been tested by numerous organizations during a limited research preview. The feedback received has led to improvements in the tool, which now offers features such as:
- Scheduled and targeted scans
- Seamless integration with audit systems
- Improved tracking of findings
Organizations can start scanning without needing API integration or custom setups, making it accessible for those already using Claude.
How it works: Users can access Claude Security directly from the Claude.ai sidebar or at claude.ai/security. To initiate a scan, users select a repository or specific directory, allowing Claude to analyze the code similarly to a security researcher. This involves understanding component interactions, tracing data flows, and interpreting source code.
Upon completion of a scan, Claude provides detailed reports on vulnerabilities, including confidence levels, severity, potential impacts, and reproduction steps. Additionally, it generates targeted patch instructions for users to implement directly.
Early adopters of Claude Security have reported significant improvements in their security workflows:
- High-confidence findings accelerate security processes.
- Reduction in time from scan to fix, with some teams applying patches in real-time.
- Ongoing coverage through scheduled scans is preferred over one-off audits.
New features in this release include the ability to target specific directories, dismiss findings with documented reasons, export results for tracking, and integrate with tools like Slack and Jira.
Organizations using Claude Security have shared positive experiences, highlighting its effectiveness in identifying vulnerabilities and streamlining their security efforts. As AI technology continues to advance, Claude Security aims to remain at the forefront of cybersecurity solutions.
Claude Security is now available for Claude Enterprise customers, with plans for access to Claude Team and Max customers in the near future. Administrators can enable this feature via the admin console, and a comprehensive Getting Started Guide is available for assistance.
