Anthropic Launches Claude Security Plugin for Claude Code in Beta
Anthropic has released the Claude Security plugin in beta for its Claude Code developer CLI tool. This plugin allows developers to scan their commits and entire codebases for security vulnerabilities directly from their terminal. By integrating vulnerability scanning directly into the terminal workflow, developers can catch and fix security issues like injection and unsafe deserialization before code reaches a pull request. This shifts security left in the development lifecycle, reducing the burden on downstream human reviewers. The plugin sends data to a model endpoint to perform reviews and covers common web vulnerability classes. It also allows developers to define custom rules specific to their codebase using a configuration file.
## BACKGROUND
Claude Code is an agentic coding tool developed by Anthropic that runs in the terminal, allowing developers to edit files, run commands, and understand codebases. Traditionally, application security testing is done via separate CI/CD pipelines or IDE extensions, whereas this tool integrates security checks directly into the CLI agent.