You can run a Workflowy MCP server locally to let AI assistants read and modify your Workflowy lists through a standard Model Context Protocol interface. This lets you search, create, update, and mark Workflowy nodes as complete or incomplete from prompts and automation, all while keeping your data under your control.

How to use

Start the server on your machine, then connect your MCP client (your AI assistant) to it. Once connected, you can use the available tools to read your Workflowy lists, search for nodes, create new nodes, update existing ones, and toggle their completion status. Use natural prompts like: find all notes under a specific project, create a new task under a milestone, or mark a set of notes as complete.

Begin by starting the MCP server from your command line. The server accepts your Workflowy credentials and then exposes a standard interface for your AI assistant to interact with. After the server is running, configure your AI assistant to point to the server's endpoint, then issue high-level requests such as listing root nodes, querying for a term, or updating a node’s text.

How to install

Prerequisites you must have before installation.

# Ensure you have Node.js v18 or higher
node -v
npm -v

"},

Install the MCP Workflowy server globally or run it via npx.

Install the package globally

npm install -g mcp-workflowy

Or run directly with npx

npx mcp-workflowy server start

## Additional setup and configuration

Create a .env file in your project directory (or export environment variables in your shell) with your Workflowy credentials.

WORKFLOWY_USERNAME=your_username_here WORKFLOWY_PASSWORD=your_password_here

## Starting the server and connecting an MCP client

To start the server using npx, run the start command. Once the server is running, configure your MCP client to connect to the server following its connection instructions.

npx mcp-workflowy server start

## Available tools

### list\_nodes

Return a list of nodes from Workflowy, either root nodes or the children of a specified node.

### search\_nodes

Search for nodes by a text query across your Workflowy structure.

### create\_node

Create a new node in Workflowy under a specified parent or root.

### update\_node

Update the text or description of an existing node in Workflowy.

### toggle\_complete

Mark a node as complete or incomplete.