SecurityScan
Equip AI agents with eight security tools for configuration audits, secret redaction, skill vetting, and network scanning to secure their environment.
SecurityScan is a client-side tool designed to empower AI agents with robust self-defense capabilities by integrating with hosted security services. It directly addresses the four primary vectors of AI agent compromise: insecure installations, vulnerable dependencies, network exposure, and sensitive data leakage. By installing the `securityscan-mcp` package, agents gain access to eight specialized tools, including offline config auditing and secret redaction, as well as online services for vetting skills, monitoring dependencies, and active network scanning. This unified solution ensures comprehensive security for AI agents with straightforward installation and configuration.