Todo MD MCP Server

@danjdewhurst

代码仓库

基本信息

分类

生产力

传输方式

stdio

作者

danjdewhurst

提交者

Daniel Dewhurst

关于 Todo MD MCP Server

An MCP (Model Context Protocol) server that provides todo list functionality backed by a simple markdown file. This server allows AI assistants to manage todo items in a standardized markdown format.

Todo MD MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "todo-md": {
      "command": "npx",
      "args": [
        "@danjdewhurst/todo-md-mcp"
      ],
      "env": {
        "TODO_FILE_PATH": "/Users/dan/Documents/todo.md"
      }
    }
  }
}

Todo MD MCP Server 工具

List all todos from the markdown file

Add a new todo item

Update an existing todo item

Delete a todo item

Remove all completed todo items

概览

What is Todo MD MCP?

An MCP server that provides todo list functionality backed by a simple markdown file. It enables AI assistants to create, read, update, and delete todo items using standard checkbox syntax in a .md file.

How to use Todo MD MCP?

Install globally via npm install -g @danjdewhurst/todo-md-mcp or clone and build locally. Add the server to an MCP client (e.g., Claude Desktop) using the npx command or path to the built file. Optionally set the TODO_FILE_PATH environment variable to specify where the todo.md file is stored. The server exposes five tools: list_todos, add_todo, update_todo, delete_todo, and clear_completed.

Key features of Todo MD MCP

Markdown-based storage using standard checkbox syntax
Full CRUD operations for todo items
Persistent unique IDs for reliable updates
MCP compliant server implementation
Fully typed TypeScript with error handling
Complete test suite with Vitest

Use cases of Todo MD MCP

AI assistant managing a user’s personal todo list via chat
Integrating task management into Claude Desktop workflows
Maintaining a simple project todo list in markdown format
Automating task tracking through MCP tools
Clearing completed items to keep the list tidy

常见问题

How do I install the server?

Install globally: `npm install -g @danjdewhurst/todo-md-mcp`. For local development, clone the repository, install dependencies, and build with `npm run build`.

How do I configure the todo file location?

Set the `TODO_FILE_PATH` environment variable in your MCP client configuration. For example: `"env": { "TODO_FILE_PATH": "/Users/yourname/Documents/todo.md" }`. Ensure the directory exists and you have write permissions.

What is the format of the todo markdown file?

Each todo uses standard markdown checkbox syntax (`- [ ]` or `- [x]`) followed by a hidden HTML comment containing a unique ID. The file also has a header (`# Todo List`) and a footer comment.

What are the available tools?

`list_todos` (list all items), `add_todo` (create a new item), `update_todo` (modify text or completion status), `delete_todo` (remove an item by ID), and `clear_completed` (remove all completed items).

What are the system requirements?

Node.js 18 or higher and an MCP-compatible client

生产力分类下的更多 MCP 服务器

Sitelas

The website builder for AI agents. Build a real website from a template in your chat, preview it, publish when you approve, then keep running it: edit pages, swap photos, and read and answer contact-form messages.

ListingGood MCP

DedeGroup

ListingGood AI Recommendation Engine — a hosted MCP server + Claude/Cursor skills that make Amazon's AI recommend your products. Free no-login scan: https://listinggood.com/scan Get API key: https://listinggood.com/deve

Snov.io MCP

Snov.io MCP is a remote MCP server available at https://snov.io. Snov.io MCP allows your AI to handle end-to-end email outreach workflows: search and verify prospect emails, build and launch automated email campaigns, t

One

Connect your agent to the apps your team already uses: Gmail, Slack, Stripe, Shopify, HubSpot, Notion, Linear, Salesforce, QuickBooks and more. One hosted MCP server with OAuth sign-in, nothing to install and no API keys

YouTube Transcript AI

youtube-transcript.ai

Fetch the full transcript of any YouTube video as clean text — no API key, no signup, no quota wall. Pass a YouTube URL (watch, youtu.be, shorts, embed) or a bare video ID and get the transcript back with title and metad