Task Manager MCP Server
Enables AI assistants to manage tasks through a comprehensive interface with 8 tools for creating, updating, searching, and tracking tasks with priorities, categories, and due dates. Features persistent file-based storage, advanced filtering, and task statistics for complete task management workflow.
README
Task Manager MCP Server
A production-ready Task Manager server built with the Model Context Protocol (MCP), enabling AI assistants to manage tasks through a standardized interface. Built with TypeScript, Zod validation, and the official MCP SDK.
Features
- ✅ 8 Comprehensive Tools: Create, list, update, delete, complete, search tasks, get statistics, and clear completed tasks
- 🔐 Type-Safe: Built with TypeScript and runtime validation using Zod
- 📦 Portable: Uses only official MCP SDK - no vendor lock-in
- 🐳 Dockerized: Ready for containerized deployment
- 💾 Persistent Storage: File-based JSON storage with environment-aware configuration
- 🔍 Advanced Filtering: Filter tasks by status, priority, and category
- 📊 Statistics & Analytics: Track task completion rates, overdue items, and more
- 🎯 Production Ready: Comprehensive error handling and validation
Quick Start
Prerequisites
- Node.js 18+
- npm 9+
Installation
# Clone the repository
git clone https://github.com/aafsar/task-manager-mcp-server.git
cd task-manager-mcp-server
# Install dependencies
npm install
# Build the project
npm run build
# Run the server
npm start
Development Mode
# Run with hot reload
npm run dev
Available Tools
1. create_task
Create a new task with optional metadata.
Parameters:
title(string, required): Task titledescription(string, optional): Detailed descriptionpriority(enum, optional): "low", "medium", or "high" (default: "medium")category(string, optional): Task category (e.g., "work", "personal")dueDate(string, optional): Due date in YYYY-MM-DD format
Example:
{
"title": "Review pull requests",
"description": "Review open PRs for the API project",
"priority": "high",
"category": "work",
"dueDate": "2025-10-05"
}
2. list_tasks
List tasks with optional filters.
Parameters:
status(enum, optional): "pending", "in_progress", "completed", or "all" (default: "all")priority(enum, optional): "low", "medium", "high", or "all" (default: "all")category(string, optional): Filter by specific category
Example:
{
"status": "pending",
"priority": "high"
}
3. update_task
Update any field of an existing task.
Parameters:
taskId(string, required): Task ID (minimum 8 characters)title(string, optional): New titledescription(string, optional): New descriptionpriority(enum, optional): New prioritycategory(string, optional): New categorydueDate(string, optional): New due datestatus(enum, optional): New status
Example:
{
"taskId": "a1b2c3d4",
"status": "in_progress",
"priority": "high"
}
4. complete_task
Mark a task as completed.
Parameters:
taskId(string, required): Task ID (minimum 8 characters)
5. delete_task
Delete a task permanently.
Parameters:
taskId(string, required): Task ID (minimum 8 characters)
6. search_tasks
Search tasks by title or description.
Parameters:
query(string, required): Search query
Example:
{
"query": "review"
}
7. get_task_stats
Get comprehensive statistics about all tasks.
Returns:
- Total task count
- Completion rate
- Status breakdown (pending/in progress/completed)
- Priority breakdown (high/medium/low)
- Category distribution
- Overdue task count
- Tasks due within 7 days
8. clear_completed
Remove all completed tasks from storage.
Claude Desktop Integration
Configuration
-
Locate your Claude Desktop config file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
- macOS:
-
Add the server configuration:
{
"mcpServers": {
"task-manager": {
"command": "node",
"args": [
"/absolute/path/to/task-manager-mcp-server/dist/index.js"
]
}
}
}
-
Restart Claude Desktop completely
-
Look for the hammer icon (🔨) in the input box
-
Test with: "Create a high priority task called 'Test MCP Integration'"
Testing with MCP Inspector
The MCP Inspector provides a web-based interface for testing tools:
# Launch the inspector
npx @modelcontextprotocol/inspector dist/index.js
This will open a browser window where you can:
- View all available tools
- Test tool execution interactively
- Inspect request/response data
- Debug errors
Docker Deployment
Build and Run with Docker
# Build the image
docker build -t task-manager-mcp .
# Run the container
docker run -it task-manager-mcp
Using Docker Compose
# Start the service
docker-compose up -d
# View logs
docker-compose logs -f
# Stop the service
docker-compose down
Persist Data with Docker
Data is automatically persisted to a Docker volume. To back up your tasks:
# Export tasks
docker cp task-manager-mcp:/app/data/tasks.json ./backup-tasks.json
# Import tasks
docker cp ./backup-tasks.json task-manager-mcp:/app/data/tasks.json
Environment Variables
Configure the server using environment variables:
# Data storage directory (default: ./data)
DATA_DIR=/custom/path/to/data
# Log level
LOG_LEVEL=info
# Node environment
NODE_ENV=production
Create a .env file in the project root:
cp .env.example .env
# Edit .env with your values
Cloud Deployment Options
Railway
# Install Railway CLI
npm install -g @railway/cli
# Login
railway login
# Initialize project
railway init
# Deploy
railway up
Render
- Connect your GitHub repository
- Create a new Web Service
- Set build command:
npm install && npm run build - Set start command:
npm start - Deploy
Fly.io
# Install flyctl
curl -L https://fly.io/install.sh | sh
# Login
flyctl auth login
# Launch app
flyctl launch
# Deploy
flyctl deploy
Project Structure
task-manager-mcp-server/
├── src/
│ ├── index.ts # Main MCP server and request handlers
│ ├── types.ts # TypeScript interfaces and Zod schemas
│ ├── storage.ts # File-based storage module
│ └── tools.ts # Tool implementation functions
├── dist/ # Compiled JavaScript (generated)
├── data/ # Task storage (JSON files, git-ignored)
├── Dockerfile # Docker configuration
├── docker-compose.yml # Docker Compose setup
├── package.json # Dependencies and scripts
├── tsconfig.json # TypeScript configuration
└── README.md # This file
Development
Scripts
npm run build # Compile TypeScript to JavaScript
npm run dev # Development mode with hot reload
npm run typecheck # Type check without building
npm run clean # Remove build artifacts
npm start # Run production build
Type Safety
The project uses strict TypeScript settings and Zod for runtime validation:
- Compile-time safety: TypeScript catches type errors during development
- Runtime validation: Zod validates all tool inputs at runtime
- Dual schema approach: JSON Schema for MCP protocol, Zod for validation
Adding New Tools
- Define Zod schema in
src/types.ts - Implement handler function in
src/tools.ts - Add tool definition to
TOOLSarray insrc/index.ts - Add case handler in
tools/callswitch statement - Rebuild and test with MCP Inspector
Troubleshooting
"Cannot find module" errors
Ensure all imports use .js extension (even for .ts files):
import { Task } from "./types.js"; // ✅ Correct
import { Task } from "./types"; // ❌ Wrong
Tasks not persisting
- Check
DATA_DIRenvironment variable - Verify write permissions on data directory
- Check for errors in server logs
TypeScript compilation errors
# Run type checker to identify issues
npm run typecheck
# Common fix: ensure strict types are used
# Check tsconfig.json module settings
MCP Inspector not connecting
- Ensure server builds successfully:
npm run build - Check Node.js version (must be 18+)
- Verify no port conflicts
- Check firewall settings
Claude Desktop not showing tools
- Verify JSON syntax in config file
- Use absolute paths in configuration
- Restart Claude Desktop completely (Cmd+R / Ctrl+R not sufficient)
- Check server logs for errors
Resources
License
MIT
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
Support
For issues and questions:
- GitHub Issues: https://github.com/aafsar/task-manager-mcp-server/issues
- MCP Discord: https://discord.gg/modelcontextprotocol
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。