reclaim-mcp-server
unofficial MCP (Model Context Protocol) server for Reclaim.ai calendar integration - manage tasks, habits, and smart scheduling through AI assistants like Claude.
README
Reclaim.ai MCP Server (UNOFFICIAL)
<!-- mcp-name: io.github.universalamateur/reclaim-mcp-server -->
UNOFFICIAL – Not affiliated with Reclaim.ai. Uses their public API , use at your own risk.
Control your Reclaim.ai calendar, tasks, and habits from AI assistants via MCP.
What You Can Do
"Create a 2-hour task to review the Q1 budget, due Friday"
"What's on my calendar tomorrow?"
"Mark my morning workout habit as done"
"Show me my productivity stats for last week"
"Start a focus block for the next 90 minutes"
40 tools across tasks, calendar, habits, focus time, and analytics. See docs/TOOLS.md for complete reference.
Quick Start
1. Get your API key: https://app.reclaim.ai/settings/developer
2. Install
pip install reclaim-mcp-server
3. Configure Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json)
{
"mcpServers": {
"reclaim": {
"command": "uvx",
"args": ["reclaim-mcp-server"],
"env": { "RECLAIM_API_KEY": "your_key_here" }
}
}
}
4. Restart Claude Desktop
Installation Options
| Method | Command |
|---|---|
| uvx (recommended) | uvx reclaim-mcp-server |
| pip | pip install reclaim-mcp-server |
| Smithery | npx -y @smithery/cli install universalamateur/reclaim-mcp-server --client claude |
| Docker | docker pull universalamateur/reclaim-mcp-server |
| Source | git clone https://gitlab.com/universalamateur1/reclaim-mcp-server.git && cd reclaim-mcp-server && poetry install |
Registries: PyPI · Smithery · Glama · GitHub · GitLab
Tool Profiles
Limit exposed tools via RECLAIM_TOOL_PROFILE:
| Profile | Tools | Use Case |
|---|---|---|
minimal |
20 | Basic task/habit management |
standard |
32 | Daily productivity |
full |
40 | All features (default) |
{
"mcpServers": {
"reclaim": {
"command": "uvx",
"args": ["reclaim-mcp-server"],
"env": {
"RECLAIM_API_KEY": "your_key_here",
"RECLAIM_TOOL_PROFILE": "standard"
}
}
}
}
Other Configurations
<details> <summary>Docker</summary>
{
"mcpServers": {
"reclaim": {
"command": "docker",
"args": ["run", "-i", "--rm", "-e", "RECLAIM_API_KEY", "universalamateur/reclaim-mcp-server"],
"env": { "RECLAIM_API_KEY": "your_key_here" }
}
}
}
</details>
<details> <summary>Poetry (from source)</summary>
{
"mcpServers": {
"reclaim": {
"command": "/opt/homebrew/bin/poetry",
"args": ["--directory", "/path/to/reclaim-mcp-server", "run", "reclaim-mcp-server"],
"env": { "RECLAIM_API_KEY": "your_key_here" }
}
}
}
Note: Use --directory flag instead of cwd – Claude Desktop doesn't respect cwd.
</details>
<details> <summary>Other MCP clients (Cursor, Continue, Zed, etc.)</summary>
Works with any MCP-compatible client. Generic stdio config:
{
"command": "uvx",
"args": ["reclaim-mcp-server"],
"env": { "RECLAIM_API_KEY": "your_key_here" }
}
</details>
Troubleshooting
| Issue | Solution |
|---|---|
| "Invalid API key" | Verify key at https://app.reclaim.ai/settings/developer |
| Rate limited (429) | Server handles retries automatically; reduce request frequency |
| Tools not showing | Restart Claude Desktop after config change |
| Docker ARM64 warning | Use v0.8.1+ for native Apple Silicon support |
Development
poetry install
poetry run pytest # Run tests
poetry run black src tests # Format
poetry run mypy src # Type check
poetry run fastmcp dev src/reclaim_mcp/server.py # Dev mode
Links
License
MIT – Falko Sieverding (@UniversalAmateur)
<sub>Built with FastMCP · Not affiliated with Reclaim.ai</sub>
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。