openclaw-twoway deployment

Deploy OpenClaw with a cloud gateway using Tailscale and SSH tunnel for secure local control, including auto environment check and firewall setup.

This package appears to implement the claimed deployment, but take precautions before running the scripts as root: 1) Inspect the scripts line-by-line (they run curl | sh and npm install -g). 2) Remove or change the insecure options (plan2 sets dangerouslyAllowInsecurePrivateWs and an environment flag OPENCLAW_ALLOW_INSECURE_PRIVATE_WS=1). 3) Don't leave generated tokens in plaintext; store them with restrictive permissions (chmod 600) or use a secret store. 4) Prefer the official npm registry if you mistrust mirrors, and audit the openclaw package code before global install. 5) Limit exposure of SSH (port 22) and gateway ports in cloud security groups—use key-based SSH and restrict source IPs where possible. 6) If unsure, run the deployment first in an isolated VM or test instance rather than a production host. If you want, I can highlight the exact lines in the scripts that set insecure flags, write tokens to files, or run remote installers so you can review them.

Type: OpenClaw Skill Name: openclaw-two-way-deployment Version: 1.0.0 The skill bundle contains multiple deployment scripts (e.g., plan1-cloud-gateway.sh, plan2-cloud-remote-gateway.sh, and plan3-local-gateway.ps1) that perform high-privilege system modifications. These include installing software globally via 'curl | sh' (Tailscale and NodeSource), modifying firewall rules (ufw, firewalld, iptables), and establishing persistence through systemd services on Linux and scheduled tasks on Windows. While these actions are consistent with the stated purpose of deploying a network gateway, the requirement for root/SYSTEM privileges and the automated modification of security configurations represent significant risky capabilities.