security-scan

To scan Claude Code configurations for security vulnerabilities, run the security-scan Skill to automatically check files like CLAUDE.md, settings.json, and MCP servers for misconfigurations and injection risks.

Injection risk assessment in Claude Code settings analyzes hooks and agent definitions to identify command injection vulnerabilities, while misconfiguration detection checks settings.json and MCP servers for improper configurations that could cause security breaches.

Yes, you need AgentShield installed as a required dependency, because the security-scan Skill relies on AgentShield to perform the underlying vulnerability scanning and assess injection risks across your Claude Code configuration files.

You can integrate Claude Code security checks into a CI/CD pipeline by running the security-scan Skill as an automated step, ensuring configurations, hooks, and agent definitions are validated for vulnerabilities before committing changes or deploying.

You should run a security scan on your Claude Code project when setting up a new environment, after modifying configuration files like CLAUDE.md or settings.json, or before committing changes to ensure your setup remains secure against vulnerabilities.

Yes, the security-scan Skill checks MCP servers and hooks for vulnerabilities, analyzing agent definitions and configurations to detect command injection risks and misconfigurations that could compromise your Claude Code environment.