WordPress launches new Claude connector for site management

On Thursday, WordPress launched a new connector for Claude, the artificial intelligence chatbot developed by Anthropic. The new integration allows WordPress site owners to share their back-end data with the chatbot system. Under this setup, users can control the specific data they want to share with the assistant, and they retain the ability to revoke this access at any time.

Currently, the integration is limited to read-only access. This restriction ensures that the Claude chatbot cannot alter, modify, or delete any information within the user’s Content Management System (CMS)—the software platform used to manage and publish website content. Instead, the chatbot can only view and analyze the permitted data. However, last year WordPress claimed that it would eventually deliver “write” access to the Model Context Protocol (MCP) integration, which is a standard designed for connecting artificial intelligence assistants to external systems. This future update would presumably allow users to conduct editorial tasks directly through a connected chatbot of their choice.

For now, the read-only connector enables site owners to query site metrics and manage comments directly through the chatbot interface using natural language. Once the system is linked, users can ask Claude to analyze back-end data, such as summarizing monthly web traffic or identifying which posts have low user engagement. To assist users, WordPress has provided template prompts for the chatbot. These templates allow users to ask questions such as which of their sites receives the most traffic, or which posts are generating the most discussion. Other functions supported by the connector include comment management, which lets users view pending comments, and plug-in management, which allows users to check what plug-ins are installed on their main site.

Why it matters

This integration bridges the gap between Content Management System back-ends and artificial intelligence assistants. By allowing site owners to query performance metrics and manage content through natural language prompts, it simplifies basic administrative workflows.