Anthropic Launches Claude Code Security for AI-Powered Vulnerability Scanning

Anthropic introduces “Claude Code Security,” a new AI feature to scan software codebases for vulnerabilities and recommend patches.
Currently available in limited research preview for Enterprise and Team customers.
The feature aims to detect vulnerabilities that traditional methods often miss by simulating the analytical capabilities of a human security researcher.
Anthropic emphasizes that the AI tool is designed to counter AI-enabled attacks by providing defenders with a more robust security baseline.
Vulnerabilities are passed through a multi-stage verification process to reduce false positives and include a severity rating for prioritized attention.
The results are presented in an analyst dashboard where humans review and approve patches, maintaining a human-in-the-loop approach.
Each identified vulnerability is given a confidence rating to assist in assessing their severity and significance.

You may have missed