Connect Claude, Cursor or Codex to WordPress. Give it the task, preview the edit, and turn important corrections into checks for the next change.
Self-hosted · MIT licensed · Bring your own AI client
Preview the selected text change.
Check the required terms are still present.
Approve, apply, and read back the result.
The requirement stays attached to the page for the next supported edit.
Content fixes, settings updates and field changes belong in the same conversation where you asked for them.
Preview a text replacement on a selected page while preserving the required booking terms. Save those terms as a check for later supported edits.
Replace a specific broken link in stored post content. Review the matching text and proposed replacement before applying it.
Inspect the site, its active plugins, options and post metadata. Identify the actual value behind the requested change.
Review the saved prior value and change record. Then run WPGuard’s desktop and mobile render check and keep the screenshot receipt.
“Keep these terms.” “Use this exact heading.” Save the requirement once, in the right scope, and check it before future supported edits.
Turn a specific requirement into a check for a named site and field. Supported edits are checked against active rules before they write.
A rule for one page stays on that page. A client preference does not become a rule for every site you manage.
Keep the originating correction alongside the rule. Review, test, or retire it when the requirement changes.
Keep a private record of requests and corrections so your assistant can retrieve earlier work for the right client.
WPGuard gives your AI a structured path from a request to a WordPress change.
Read the site context and the value you plan to change. Choose a supported action for that site and transport.
See the previous and proposed value. Review the exact target and any applicable correction checks.
Approve the change packet, then apply the matching request. Pass the preview’s state check to catch edits made in the meantime.
Read the stored result back, then capture desktop and mobile screenshots with checks for expected text, overflow, headings, and broken images.
The open-source server runs on your infrastructure and connects over SSH with WP-CLI or an HTTPS companion plugin. Your AI client must be able to reach its authenticated MCP endpoint.
Local site connections, supported edits, change records and correction checks. No Cloud subscription required.
A separate workspace for paired instances, packet history and approval decisions. New organizations need access arranged with us. See the published offers →
Connect one site. Inspect it. Preview one useful change.