zaherg/wp-blockmarkup-mcp-server — explained in plain English
Analysis updated 2026-05-18
Let an AI coding assistant look up real Gutenberg block schemas before generating markup.
Index WordPress core, WooCommerce, or a custom plugin's blocks for search and validation.
Validate existing block markup against real attribute and nesting rules.
Host a shared, authenticated MCP instance so a whole team can query indexed blocks.
| zaherg/wp-blockmarkup-mcp-server | 0xmukesh/docusaurus-tutorial | 1tsmejp/palworld-docker-wine | |
|---|---|---|---|
| Stars | 1 | 1 | 1 |
| Language | JavaScript | JavaScript | JavaScript |
| Last pushed | — | 2021-12-27 | — |
| Maintenance | — | Dormant | — |
| Setup difficulty | moderate | easy | moderate |
| Complexity | — | 2/5 | 4/5 |
| Audience | developer | developer | ops devops |
Figures from each repo's GitHub metadata at analysis time.
wp-blockmarkup-mcp-server is a tool that helps AI coding assistants generate correct WordPress content blocks, known as Gutenberg blocks, instead of guessing at them. It is an experimental fork and refactor of an earlier project, and its author is upfront that it is early stage and that anything generated with it should be checked in a real WordPress environment before being trusted in production. The core problem it solves is that AI assistants often try to write Gutenberg block markup from memory, which can be wrong or outdated. They might invent attributes that do not exist, use incorrect styling conventions, ignore how blocks are supposed to nest inside each other, or produce plain static HTML for blocks that are actually meant to be dynamic. WordPress then rejects this content as invalid. This project fixes that by actually indexing the real source code of WordPress itself, WooCommerce, or any other block based plugin, so an AI assistant can look up how a block is truly defined, including its attributes, supported features, variations, and how it should be saved, before generating any markup. It works using the Model Context Protocol, or MCP, which lets compatible AI tools like Claude Code connect to it as a server. You run it locally by cloning the code and using Node.js, and you can add different sources to index, such as a public GitHub repository, a private one using an access token, or a local plugin folder you are actively developing. Once a source is indexed, you can search for blocks, look up a block's structure, and validate markup you already have against real rules, all from a simple command line interface. Compared to the original project it is based on, this fork adds a way to run it as a shared online service using Docker, with authentication, safety checks around which GitHub repositories and paths it is allowed to touch, and persistent storage using SQLite, so a whole team can use one hosted instance instead of everyone running their own local copy. This tool is aimed at developers building WordPress sites or plugins with the help of AI coding assistants, especially teams who want their AI tools to generate Gutenberg blocks that actually work the first time.
An MCP server that indexes real WordPress Gutenberg block source code so AI assistants generate valid block markup instead of guessing.
Mainly JavaScript. The stack also includes JavaScript, Node.js, MCP.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
double-check against the repo, no cap.