Todoist MCP Server
Enables task, project, section, label, and comment management via Todoist's API. Supports batch operations and pagination.
README
Todoist MCP Server
MCP server for Todoist API integration. Published to PyPI
Status
✅ Full Functionality - Now using Todoist unified API v1 with proper pagination support
v0.4.0 - Major expansion: Sections support, Labels management, and Batch operations
Important API Notes
Todoist has numerous API versions with incompatible ID systems:
- v1 Unified API (Todoist official ): Uses alphanumeric IDs (e.g.,
69mF7QcCj9JmXxp8) - v2 REST API: Uses numeric IDs (e.g.,
7246645180)
This incompatibility means certain features like server-side search are not available in this MCP implementation, as search only exists in v2 but returns IDs incompatible with v1 operations.
Features
- Full cursor-based pagination for all endpoints
- Configurable limit parameter for all list endpoints
- Multi-auth support (environment, config file, runtime)
- Complete task and project management capabilities
- Sections support (v0.4.0): CRUD operations for project sections
- Labels management (v0.4.0): CRUD operations for labels
- Batch operations (v0.4.0): Move, update, complete multiple tasks at once
- Comment CRUD operations for tasks and projects
- Move tasks between projects, sections, and parents
- Error handling with detailed error messages
Limitations
- No server-side search (due to v1/v2 API incompatibility)
- Client-side filtering only for task queries
Installation
pip install todoist-mcp
Claude Desktop Configuration
Add to your claude_desktop_config.json:
{
"mcpServers": {
"todoist": {
"command": "uvx",
"args": ["todoist-mcp"],
"env": {
"TODOIST_API_TOKEN": "your_token_here"
}
}
}
}
AmazonQ Developer CLI Configuration
Add to your .amazonq/mcp.json:
{
"mcpServers": {
"todoist": {
"command": "uvx",
"args": ["todoist-mcp"],
"env": {
"TODOIST_API_TOKEN": "your_token_here"
}
}
}
}
Alternative: Using mcp-gen
For complex MCP configuration management such as in providing configurations for different contexts ot teams, use mcp-gen with YAML:
See https://github.com/OAuthBringer/mcp-gen/examples for more full examples
# mcp-config.yaml
servers:
todoist:
command: uvx
args: [todoist-mcp]
env:
TODOIST_API_TOKEN: {{ secrets.TODOIST_API_TOKEN }}
```
Then generate your configuration:
```bash
`mcp-gen generate -c mcp-config.yaml -s secrets-file.yaml`
Configuration
Authentication options (in order of precedence):
- Runtime: Pass token when creating client
- Config file:
~/.config/todoist/config.jsonwith{"api_token": "your_token"} - Environment: Set
TODOIST_API_TOKEN
Available Tools
Projects
get_projects- List projects with pagination (limit, cursor)get_project- Get single project by IDadd_project- Create new project
Tasks
get_tasks- List tasks with pagination and filtersget_task- Get single task by IDadd_task- Create new task with all propertiesupdate_task- Update existing taskmove_task- Move task to different project, section, or parent
Sections (v0.4.0)
get_sections- List sections for a project with paginationget_section- Get single section by IDadd_section- Create new section in projectupdate_section- Update section namedelete_section- Delete a section
Labels (v0.4.0)
get_labels- List all labels with paginationget_label- Get single label by IDadd_label- Create new labelupdate_label- Update label propertiesdelete_label- Delete a label
Batch Operations (v0.4.0)
batch_move_tasks- Move multiple tasks to project/sectionbatch_update_labels- Add/remove labels from multiple tasksbatch_update_tasks- Update multiple tasks with same propertiesbatch_complete_tasks- Complete multiple tasks at once
Comments
get_comments- List comments for task/project with paginationget_comment- Get single comment by IDadd_comment- Add comment to task or projectupdate_comment- Update existing commentdelete_comment- Delete a comment
Technical Details
- Built with FastMCP v2.3.3+
- Python 3.11+
- Direct API v1 integration using httpx
- No dependency on todoist-api-python SDK
Development
# Clone repository
git clone https://github.com/OAuthBringer/todoist-mcp
cd todoist-mcp
# Create virtual environment
uv venv
source .venv/bin/activate # or .venv\Scripts\activate on Windows
# Install dependencies
uv pip install -e .
uv pip install -e ".[dev]"
# Run tests
pytest
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。