wp-devdocs-mcp

Give your AI coding assistant a verified WordPress hook database instead of letting it guess.

wp-devdocs-mcp is a local MCP server that indexes every action, filter, block registration, and JS API call from WordPress, WooCommerce, Gutenberg, or any plugin you work with. It gives AI tools like Claude Code a verified database to search and validate against instead of relying on training data.

Why This Exists

AI coding assistants are changing how we build WordPress plugins. Tools like Claude Code, Cursor, and Windsurf can scaffold entire plugins in minutes — but they all share the same blind spot: hook names come from training data, not from the actual source code.

Most of the time this works fine. Models like Claude Sonnet nail common hooks almost every time. But "most of the time" isn't good enough when you're shipping production code, and not every model is Claude Sonnet. Across the landscape of LLMs doing code generation today:

• Some models invent hooks that don't exist — woocommerce_email_after_order_details sounds right but isn't real

• Some use deprecated namespaces — wp.editor.InspectorControls instead of wp.blockEditor.InspectorControls

• Most miss newer hooks — suggesting pre_get_posts for WooCommerce order queries when HPOS uses woocommerce_order_list_table_prepare_items_query_args

• Parameters get mixed up — wrong argument count or order in callback signatures

• No model knows your custom plugins — private or proprietary hooks are invisible to every model's training data

The core issue is simple: we can't rely 100% on any model to produce correct WordPress hook names from memory alone. Even the best models benefit from verification, and the rest genuinely need it. You only find out about hallucinated hooks when the code doesn't work — and in an agentic workflow where the AI writes, tests, and iterates autonomously, one bad hook name can send it down a rabbit hole of debugging something that was never going to work.

The Solution

Feed the LLM real data. Instead of hoping the model remembers the right hook name, give it a verified database to query.

wp-devdocs-mcp parses the actual source code of any WordPress plugin and builds a searchable index of every hook with its exact name, type, parameters, file location, and surrounding code context. Your AI assistant queries this index before writing code — so every hook name in the generated code is verified against the real source.

This fits naturally into agentic coding workflows. When Claude Code (or any MCP-compatible assistant) needs to use a WordPress hook, it:

1. Searches the indexed database for relevant hooks

2. Validates the exact hook name exists before writing it into code

3. Reads the context — parameters, docblock, surrounding code — to use it correctly

No hallucination. No guessing. No debugging phantom hooks.

What gets indexed:

What the AI gets for each hook:

• Exact name (with dynamic name detection for hooks like woocommerce_thankyou_{$payment_method})

• Type (action / filter / js_action / js_filter)

• Parameters and count

• File path and line number

• Enclosing function and class

• Docblock

• Code window (8 lines before, 4 after)

• Source plugin name

Quick Start

Index Your First Source

Each source:add command clones the repo and indexes it automatically:

That's it. 3,500+ hooks indexed in under a minute.

Connect to Claude Code

Add the MCP server to your Claude Code configuration. Create or edit .mcp.json in your project root:

Now when you ask Claude Code to write WordPress plugin code, it will automatically search and validate hook names against your indexed sources before generating code.

Indexing Sources

WordPress Core

Expected output:

Gutenberg

WooCommerce

Any Public Plugin

Your Private Plugins

For private GitHub repos, store your token in an environment variable (never in the config):

Local Plugin Development

Point directly at a folder on your machine — great for plugins you're actively developing:

Source Options

CLI Reference

CLI Examples

MCP Tools

When connected to Claude Code (or any MCP-compatible client), four tools are available:

search_hooks

Full-text search with BM25 ranking across all indexed hooks. Supports filters for type, source, dynamic hooks, and removed hooks.

validate_hook

Exact-match check — returns VALID with file locations, NOT_FOUND with similar suggestions, or REMOVED for deprecated hooks. This is how the AI confirms a hook name before using it in code.

get_hook_context

Returns the full code window around a hook: the line itself, 8 lines before, 4 lines after, the docblock, enclosing function, and class. Gives the AI enough context to use the hook correctly.

search_block_apis

Searches block registrations (registerBlockType, etc.) and JavaScript API usages (wp.blockEditor.*, wp.data.*, etc.). Only matches on structured fields (block name, API call, namespace) — not surrounding code — to prevent false positives.

How It Works

1. Sources are registered via the CLI — each points to a GitHub repo or local folder

2. Indexing clones/pulls the repo, scans PHP and JS/TS files, and extracts hooks using regex-based parsers

3. Storage uses SQLite with FTS5 full-text search and WAL mode for fast concurrent reads

4. Incremental updates skip files that haven't changed (mtime + content hash)

5. Soft-delete tracking marks hooks that were previously indexed but no longer found as removed

6. The MCP server exposes the database as tools over stdio — your AI assistant queries it in real-time

Data Storage

All data lives in ~/.wp-devdocs-mcp/:

Requirements

• Node.js 20+

• ~500MB disk space per large plugin source (WooCommerce, Gutenberg)

License

MIT