Super Productivity MCP Server
Enables Claude to interact with Super Productivity's local REST API for task, project, and tag management, including CRUD operations, status checks, and daily planning prompts.
README
Super Productivity — MCP Server
Integrates Claude with Super Productivity using the Model Context Protocol.
This server targets the desktop app's Local REST API on http://127.0.0.1:3876.
Only tested on macOS. Contributions and bug reports are welcome!
Structure
src/
├── index.ts ← entry point, starts the MCP server
├── sp-client.ts ← wrapper for Super Productivity REST API
├── resources.ts ← resources readable by Claude (tasks, projects, tags)
├── prompts.ts ← reusable prompt templates
└── tools/
├── tasks.ts ← CRUD tools for tasks
└── projects.ts ← tools for projects and tags
Setup
1. Install dependencies
npm install
2. Build
npm run build
# or use watch mode during development:
npm run dev
3. Enable the API in Super Productivity
In Super Productivity → Settings → Misc → Enable local REST API.
The desktop app listens on 127.0.0.1:3876.
4. Register in Claude Desktop
Open ~/Library/Application Support/Claude/claude_desktop_config.json (macOS)
or %APPDATA%\Claude\claude_desktop_config.json (Windows)
and add the block from claude_desktop_config.snippet.json,
replacing /ABSOLUTE/PATH/ with the real project path:
{
"mcpServers": {
"super-productivity": {
"command": "/usr/local/bin/node",
"args": ["/absolute/path/super-productivity-mcp/dist/index.js"],
"env": {
"SP_API_URL": "http://127.0.0.1:3876"
}
}
}
}
Restart Claude Desktop. You should see the 🔧 icon in chat — this means the tools are active.
4a. Quick smoke test
Before opening Claude, verify the local API:
curl http://127.0.0.1:3876/health
curl http://127.0.0.1:3876/tasks
5. (Optional) Debug with MCP Inspector
npm run inspect
This opens a web interface for manually testing each tool.
Available Tools
| Tool | Description |
|---|---|
health_check |
Checks that the local REST API is up and ready |
list_tasks |
Lists tasks with optional filters (query, projectId, projectName, tagId, tagName, includeDone, source) |
search_tasks |
Searches tasks by title text with the same optional filters |
get_task |
Retrieves the details of a single task |
get_status |
Returns current status: active task and task counts |
get_current_task |
Returns the currently running task |
create_task |
Creates a new task; accepts projectName and tagNames in addition to IDs |
update_task |
Updates fields of an existing task; also supports projectName and tagNames |
complete_task |
Marks a task as completed |
start_task |
Starts a task as the current task |
set_current_task |
Sets or clears the current task |
stop_current_task |
Stops the current task |
archive_task |
Archives a task |
restore_task |
Restores an archived task |
delete_task |
Deletes a task |
get_projects |
Lists projects |
get_tags |
Lists tags |
Available Resources
| URI | Description |
|---|---|
sp://status |
Current status in text format |
sp://tasks/active |
Active tasks in text format |
sp://tasks/archived |
Archived tasks in text format |
sp://task/current |
Current task in JSON |
sp://projects |
All projects in JSON |
sp://tags |
All tags in JSON |
Available Prompts
| Name | Description |
|---|---|
daily-planning |
Daily planning session |
weekly-review |
Weekly review |
break-down-task |
Breaks down a complex task into subtasks |
Environment Variables
| Variable | Default | Description |
|---|---|---|
SP_API_URL |
http://127.0.0.1:3876 |
Base URL for the Super Productivity API |
Development
npm test
This runs the TypeScript build and the Node test suite.
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。