Anthropic Launches Claude Security for Code Vulnerability Scanning

Severity: Low (Score: 27.9)

Sources: Thenewstack, claude.com, www.wired.com

Summary

On May 1, 2026, Anthropic announced the public beta release of Claude Security, a tool designed to scan codebases for vulnerabilities and suggest fixes. Initially available only to Enterprise customers, the tool aims to address vulnerabilities that traditional scanners often miss by analyzing data flows and understanding context. Claude Security employs a multi-agent approach to scan entire codebases and includes an adversarial verification process to reduce false positives. Users can review and approve suggested patches, enabling quicker remediation of high-severity vulnerabilities. The tool is particularly effective against complex issues such as memory corruption and injection flaws. Claude Security is expected to enhance security workflows by integrating with existing tools like Slack and Jira. Future access for Team and Max plan users is anticipated soon. Key Points: • Claude Security is now in public beta for Enterprise customers. • The tool scans codebases for vulnerabilities that traditional scanners often miss. • It includes an adversarial verification process to minimize false positives.

Key Entities

• GitHub (platform)