Todoist MCP Helper
Connect AI agents to your Todoist tasks via the Model Context Protocol (MCP).
README
<p align="center"> <img src="icon.png" width="120" alt="Todoist MCP Helper Icon"> </p>
Todoist MCP Helper ✅
English | 中文
Connect AI agents to your Todoist tasks via the Model Context Protocol (MCP).
Create, search, complete, and manage your Todoist tasks — all from Claude, Gemini, Cursor, or any MCP-compatible AI agent.
✨ Features
| Category | Tools | Description |
|---|---|---|
| 📋 Tasks | get_tasks, get_task, create_task, update_task, close_task, delete_task, reopen_task |
Full task CRUD with priority, due dates, labels |
| 🚚 Move | move_task, move_task_by_name |
Move tasks across projects, sections, parents |
| 🔍 Smart Search | search_task_by_name, complete_task_by_name, delete_task_by_name, update_task_by_name |
Find and operate on tasks by name (fuzzy matching) |
| 📁 Projects | list_projects, create_project, update_project, delete_project, get_project_overview |
Manage projects + full structure overview |
| 📑 Sections | list_sections, create_section, update_section, delete_section |
Organize and rename sections |
| 🏷️ Labels | list_labels, create_label |
Tag management |
| 💬 Comments | get_comments, create_comment |
Task & project comments |
| ⚙️ Config | set_api_token, get_current_config |
Runtime token management |
27 tools total — the most comprehensive Todoist MCP server available.
🚀 Quick Start
Option A: Quick Run with uvx (Recommended)
No installation needed — runs in a temporary isolated environment:
uvx todoist-mcp-helper
Don't have
uv? Install it:curl -LsSf https://astral.sh/uv/install.sh | sh
Option B: Install with pip
pip install todoist-mcp-helper
Get Your API Token
- Go to Todoist Settings → Integrations
- Scroll to Developer → copy your API Token
📋 Configuration
All credentials are passed via environment variables — no tokens in code.
| Variable | Description | Required |
|---|---|---|
TODOIST_API_TOKEN |
Your Todoist API Token | ✅ |
🔧 Platform Configuration
Claude Desktop
Add to claude_desktop_config.json:
<table><tr><th>uvx (recommended)</th><th>pip</th></tr><tr><td>
{
"mcpServers": {
"todoist": {
"command": "uvx",
"args": ["todoist-mcp-helper"],
"env": {
"TODOIST_API_TOKEN": "your_token"
}
}
}
}
</td><td>
{
"mcpServers": {
"todoist": {
"command": "todoist-mcp-helper",
"env": {
"TODOIST_API_TOKEN": "your_token"
}
}
}
}
</td></tr></table>
Gemini CLI
Add to ~/.gemini/settings.json:
<table><tr><th>uvx (recommended)</th><th>pip</th></tr><tr><td>
{
"mcpServers": {
"todoist": {
"command": "uvx",
"args": ["todoist-mcp-helper"],
"env": {
"TODOIST_API_TOKEN": "your_token"
}
}
}
}
</td><td>
{
"mcpServers": {
"todoist": {
"command": "todoist-mcp-helper",
"env": {
"TODOIST_API_TOKEN": "your_token"
}
}
}
}
</td></tr></table>
Cursor
Add to .cursor/mcp.json:
<table><tr><th>uvx (recommended)</th><th>pip</th></tr><tr><td>
{
"mcpServers": {
"todoist": {
"command": "uvx",
"args": ["todoist-mcp-helper"],
"env": {
"TODOIST_API_TOKEN": "your_token"
}
}
}
}
</td><td>
{
"mcpServers": {
"todoist": {
"command": "todoist-mcp-helper",
"env": {
"TODOIST_API_TOKEN": "your_token"
}
}
}
}
</td></tr></table>
CherryStudio / Other MCP Clients
{
"todoist": {
"command": "uvx",
"args": ["todoist-mcp-helper"],
"env": {
"TODOIST_API_TOKEN": "your_token"
}
}
}
ModelScope (魔塔)
Select Stdio mode in the MCP service configuration, then use:
{
"mcpServers": {
"todoist": {
"command": "uvx",
"args": ["todoist-mcp-helper"],
"env": {
"TODOIST_API_TOKEN": "your_token"
}
}
}
}
Add TODOIST_API_TOKEN in the environment variables section with your API token value.
💡 Usage Examples
Once configured, ask your AI agent:
- "Show me my tasks for today"
- "Create a task: Buy groceries, due tomorrow, priority 2"
- "Complete the task about groceries"
- "Move the grocery task to my Shopping project"
- "Search for tasks related to meeting"
- "List all my projects"
- "Show me the full project overview with sections"
- "Add a comment to my latest task"
🔐 Runtime Configuration
Change tokens without restarting:
set_api_token— Switch Todoist account at runtimeget_current_config— Check current configuration
💖 Support
If this project helps you, consider buying me a coffee! Your support keeps this project alive and growing ✨
<table> <tr> <td align="center" width="50%">
☕ 爱发电 (Afdian)
<a href="https://afdian.com/a/LittlePeter52012"> <img src="https://img.shields.io/badge/爱发电-946CE6?style=for-the-badge&logo=data:image/svg+xml;base64,PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCAyNCAyNCIgZmlsbD0id2hpdGUiPjxwYXRoIGQ9Ik0xMiAyMS4zNWwtMS40NS0xLjMyQzUuNCAxNS4zNiAyIDEyLjI4IDIgOC41IDIgNS40MiA0LjQyIDMgNy41IDNjMS43NCAwIDMuNDEuODEgNC41IDIuMDlDMTMuMDkgMy44MSAxNC43NiAzIDE2LjUgMyAxOS41OCAzIDIyIDUuNDIgMjIgOC41YzAgMy43OC0zLjQgNi44Ni04LjU1IDExLjU0TDEyIDIxLjM1eiIvPjwvc3ZnPg==&logoColor=white" alt="Support on Afdian"> </a>
<sub>Supports Alipay & WeChat Pay</sub>
</td> <td align="center" width="50%">
💎 Crypto (USDC / ERC-20)
<a href="https://littlepeter52012.github.io/todoist-mcp-helper/donate.html"> <img src="https://img.shields.io/badge/USDC%2FETH-Donate_Now-6c5ce7?style=for-the-badge&logo=ethereum&logoColor=white" alt="Donate Crypto"> </a>
<sub>Click to donate — supports MetaMask, SafePal & more</sub>
</td> </tr> </table>
Every little bit helps — thank you! 🙏
📄 License
MIT License — see LICENSE for details.
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。