WordPress Self-Hosted

Manage a self-hosted WordPress site via SSH+WP-CLI (primary) and WP REST API (when direct HTTPS access is available). Use when asked to write, draft, publish...

This skill appears to do what it says, but pay attention to operational security before installing: - Pre-populate known_hosts or avoid accept-new: remove -o StrictHostKeyChecking=accept-new and add the host key to known_hosts to avoid TOFU risks. - Protect app passwords: using op is good, but be aware that passing credentials on command lines (curl -u "user:pass") or storing them in environment variables can expose them in process listings or logs on some systems; prefer using stdin, files with restrictive perms, or authenticated HTTP headers when possible. - Remote /tmp use: SCPing files to /tmp on the remote machine is convenient but can be risky on multi-tenant systems; ensure the SSH account and remote /tmp permissions are appropriate and that files are removed promptly (the instructions do this, but verify it in practice). - Least privilege SSH user: give the WP SSH user only the permissions it needs (wp-cli operations) and avoid using root or overly privileged accounts. - 1Password/SSH agent: ensure your SSH agent socket and 1Password agent are properly secured and that PTY requirements are acceptable for automation. - Test in a safe environment first: validate the flow on a staging site and confirm REST auth behavior (proxies, Cloudflare, and Wordfence may block app passwords) before running on production. If you cannot or will not accept the TOFU pattern, remote /tmp writes, or the credential handling model described, do not enable the skill until those operational details are addressed.

Type: OpenClaw Skill Name: wordpress-selfhosted Version: 1.0.3 The skill bundle provides the agent with high-risk capabilities, including remote shell access via SSH and file transfers via SCP to manage a WordPress installation. While these actions are aligned with the stated purpose, the instructions in SKILL.md include risky practices such as using 'StrictHostKeyChecking=accept-new' (TOFU) and revealing credentials via the 1Password CLI ('op item get --reveal'). The use of shell command templates also introduces a potential surface for shell injection if the agent handles unsanitized input during execution.