Personal Task Manager MCP
Enables AI assistants to manage tasks with full lifecycle support including due dates, priorities, tags, subtasks, and project lists via 19 SQLite-backed MCP tools.
README
Personal Task Manager — MCP Server
A Model Context Protocol (MCP) server that turns any AI assistant into a powerful task management system. Built with Python and SQLite, it exposes 19 tools for full lifecycle task management from quick to-dos to multi project workflows with priorities, tags, subtasks, and more.
AI Assistant <--> MCP Protocol (stdio) <--> Task Manager Server <--> SQLite DB
Features
| Feature | Description |
|---|---|
| Due Dates & Deadlines | Set, update, or clear deadlines on any task |
| Priorities | Classify tasks as high, medium, or low |
| Status Workflow | Full lifecycle: todo > in_progress > blocked > done > archived |
| Tags & Labels | Flexible many-to-many tagging (e.g. @work, @home, urgent) |
| Notes & Descriptions | Rich markdown notes attached to any task |
| Subtasks & Checklists | Break tasks into ordered, toggleable checklist items |
| Lists & Projects | Organize tasks by context — work, personal, side-project |
All 19 MCP Tools
Core CRUD
| Tool | Signature | Description |
|---|---|---|
get_schema |
() |
Get the full database schema |
run_select_query |
(query) |
Run a read-only SQL SELECT query |
add_task |
(title, description?, due_date?, priority?, list_id?, notes?) |
Create a new task |
update_task |
(task_id, title?, description?, status?, due_date?, priority?, list_id?, notes?) |
Update any task field |
delete_task |
(task_id) |
Delete a task and its subtasks/tags |
Due Dates
| Tool | Signature | Description |
|---|---|---|
set_due_date |
(task_id, due_date) |
Set or clear a deadline (YYYY-MM-DD) |
Priorities
| Tool | Signature | Description |
|---|---|---|
set_priority |
(task_id, level) |
Set priority: high / medium / low |
Status Workflow
| Tool | Signature | Description |
|---|---|---|
set_status |
(task_id, status) |
Set status: todo / in_progress / blocked / done / archived |
Tags & Labels
| Tool | Signature | Description |
|---|---|---|
add_tag |
(task_id, tag) |
Add a tag to a task (creates if new) |
remove_tag |
(task_id, tag) |
Remove a tag from a task |
list_by_tag |
(tag) |
List all tasks with a given tag |
Notes
| Tool | Signature | Description |
|---|---|---|
update_notes |
(task_id, markdown_text) |
Set/replace markdown notes on a task |
Subtasks & Checklists
| Tool | Signature | Description |
|---|---|---|
add_subtask |
(parent_id, title) |
Add a checklist item to a task |
toggle_subtask |
(subtask_id) |
Toggle completion (done / not done) |
delete_subtask |
(subtask_id) |
Remove a subtask |
list_subtasks |
(parent_id) |
List all subtasks of a task |
Lists & Projects
| Tool | Signature | Description |
|---|---|---|
create_list |
(name, description?) |
Create a new list/project |
delete_list |
(list_id) |
Delete a list (tasks become unassigned) |
move_task_to_list |
(task_id, list_id) |
Move a task to a list (use 0 to unassign) |
list_tasks_in_list |
(list_id) |
List all tasks in a list |
Quick Start
Prerequisites
- Python 3.10+
- An MCP-compatible client (e.g. Claude Desktop, Cursor, VS Code + Copilot)
Installation
# Clone the repository
git clone https://github.com/arghyapolley/personal-task-manager-mcp.git
cd personal-task-manager-mcp
# Install dependencies
pip install -r requirements.txt
Run the Server
python server.py
The server communicates via stdio using the MCP protocol. It automatically creates/migrates the tasks.db SQLite database on startup.
Connect to Claude Desktop
Add this to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"task-manager": {
"command": "python",
"args": ["/absolute/path/to/server.py"]
}
}
}
Then restart Claude Desktop. You can now say things like:
"Add a high-priority task called 'Deploy v2' to my Work project, due May 10th"
"Show me all tasks tagged @urgent"
"Add subtasks to task 3: write tests, update docs, deploy"
"What's overdue?"
Database Schema
+---------------+ +---------------+
| lists | | tags |
+---------------+ +---------------+
| id (PK) | | id (PK) |
| name (UQ) | | name (UQ) |
| description | +-------+-------+
| created_at | |
+-------+-------+ |
| +------+------+
| 1:N | task_tags |
| +-------------+
+-------+-------+ | task_id(FK) |--+
| tasks | | tag_id(FK) | |
+---------------+ +-------------+ |
| id (PK) |<-----------------------+
| title |
| description | +-------------+
| status | | subtasks |
| due_date | +-------------+
| priority | 1:N | id (PK) |
| list_id (FK) |<------| parent_id |
| notes | | title |
| created_at | |is_completed |
| updated_at | | sort_order |
+---------------+ | created_at |
+-------------+
Testing
Run the full test suite (63 tests, uses a temporary DB):
python test_server.py
Results: 63 passed, 0 failed
Tests cover: schema creation, CRUD, priorities, status workflow, tags (add/remove/list/duplicates), notes, subtasks (add/toggle/delete/ordering), lists (create/move/unassign/delete), cascade deletes, and SQL injection guards.
Project Structure
personal-task-manager-mcp/
|-- server.py # MCP server -- 19 tools, DB init, schema migration
|-- test_server.py # Comprehensive test suite (63 assertions)
|-- requirements.txt # Python dependencies (mcp, pydantic)
|-- tasks.db # SQLite database (auto-created on first run)
+-- README.md # This file
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。