What problem does it solve? Running an always-on personal AI assistant on your own hardware exposes the host to risks like open ports, weak SSH configurations, missing disk encryption, and unpatched systems. This Skill performs a structured, read-only security audit of an OpenClaw host and proposes staged, reversible hardening steps without breaking your access. ## Core Features & Use Cases - Read-only security checks: Runs commands like openclaw security audit --deep, firewall status, listening-port scans, and disk encryption checks on macOS, Linux, and Windows. - Risk profiling: Infers context (OS, access path, network exposure, backup status) and lets you choose a Convenience, Balanced, or Strict security posture. - Staged hardening plan: Produces a severity-ranked findings report with reversible recommendations such as binding the gateway to loopback, requiring auth, enabling updates, and verifying backups. - Use Case: You just deployed OpenClaw on a remote VPS. Use this Skill to audit SSH configuration, confirm the gateway is not publicly exposed, verify automatic security updates, and get a safe step-by-step lockdown plan. ## Quick Start Ask the assistant to run a security healthcheck on my OpenClaw host and recommend hardening steps.
