WordPress Expert

Enable WordPress superpowers for OpenClaw. Your Developer, Content Manager, Author, Security Specialists, Contributor, Subscriber and Admin and more.

This skill is internally consistent and appears to be what it says: a WordPress management guide for OpenClaw that can use REST, WP‑CLI, or an optional companion plugin for safer tooling. Before installing or enabling anything: 1) Do NOT paste application passwords into chat or commit them to Git—store them in environment or openclaw.json only. 2) Review the companion plugin repository (github.com/realM1lF/openclaw-wordpress-tool) and any MU plugin PHP source before running npm install or copying files to wp-content/mu-plugins — treat this as a third‑party package. 3) Use a least‑privilege WordPress account (or staging) when granting access. 4) Carefully configure tools.allow / sandbox policies so only the minimal wordpress-site-tools entries are permitted; avoid broad WP‑CLI allowlists on production. 5) Verify that the ClawHub/registry bundle did not silently omit or mutate required server-side PHP (the README warns that packaged text bundles may omit binary/.php files); clone the full repo for the mu-plugin if you intend to deploy it. If you need help reviewing the companion plugin code or deciding a safe allowlist for WP‑CLI, get that code inspected before enabling it in production.

Type: OpenClaw Skill Name: wordpress-expert Version: 1.0.12 The wordpress-expert skill bundle is a highly detailed and professional set of instructions and reference documents for managing WordPress sites via an OpenClaw agent. It prioritizes security through extensive documentation on least-privilege access, credential hygiene (using Application Passwords and environment variables), and defensive coding practices (sanitization and escaping). The bundle explicitly warns against echoing secrets and provides a safety-oriented 'Read-Plan-Write-Verify' workflow. It recommends a companion plugin (github.com/realM1lF/openclaw-wordpress-tool) to implement allowlisted WP-CLI and REST tools, reducing the risk associated with raw shell execution. No indicators of data exfiltration, backdoors, or malicious prompt injection were found across the analyzed files (SKILL.md, SAFETY.md, PRE_INSTALL_AND_TRUST.md).