Anthropic has introduced a new cybersecurity-focused product called Claude Security, designed to help organizations identify and manage vulnerabilities in their software systems. The tool is currently available in public beta for Enterprise-tier users of Claude, with plans to extend access to Team and Max-tier customers in the near future.
Claude Security is positioned as part of Anthropic’s broader push into defensive cybersecurity solutions. The system leverages the Claude Opus 4.7 model to scan large codebases, detect potential security flaws, and even suggest targeted patches. This approach is intended to support security teams by reducing manual effort and improving the speed at which vulnerabilities are identified and addressed.
According to the company, the tool not only highlights security weaknesses but also helps prioritize which issues should be fixed first. This prioritization is aimed at helping developers focus on the most critical risks, rather than being overwhelmed by large volumes of flagged issues in complex software environments.
Alongside this launch, Anthropic has also been expanding its cybersecurity research efforts through initiatives like Project Glasswing, which focuses on finding vulnerabilities in widely used open-source infrastructure. The project uses a specialized model called Mythos, which is not publicly released due to its advanced capabilities. Participants in the initiative include major industry players such as Amazon Web Services, Microsoft, Google, and others, highlighting growing collaboration across the tech industry in strengthening global software security.