Obsidian Todos MCP Server
Enables AI assistants to manage tasks within an Obsidian vault by listing, adding, and updating todos via the Local REST API. It allows users to create new todos in daily notes and retrieve task statistics through natural language.
README
Obsidian Todos MCP Server
Model Context Protocol (MCP) server that provides AI assistants with direct access to Obsidian todos through the Local REST API.
Features
- 🔍 List todos - Get all incomplete todos from your vault
- ➕ Add todos - Create new todos in today's daily note
- ✏️ Update todos - Modify existing todos or mark them complete
- 📊 Get stats - View todo statistics by file
Prerequisites
-
Obsidian with the following plugins installed and enabled:
- Local REST API
- Dataview
- Todos REST API (this plugin)
-
Node.js 18+ installed
Installation
From npm (coming soon)
npm install -g obsidian-todos-mcp-server
From source
cd mcp-server
npm install
npm run build
Configuration
Environment Variables
OBSIDIAN_API_URL- Base URL for the REST API (default:http://localhost:27124)OBSIDIAN_API_KEY- API key if authentication is enabled
Claude Desktop Configuration
Add to your claude_desktop_config.json:
{
"mcpServers": {
"obsidian-todos": {
"command": "npx",
"args": ["-y", "obsidian-todos-mcp-server"],
"env": {
"OBSIDIAN_API_URL": "http://localhost:27124"
}
}
}
}
Or use the direct binary name after npm link:
{
"mcpServers": {
"obsidian-todos": {
"command": "obsidian-todos-mcp",
"env": {
"OBSIDIAN_API_URL": "http://localhost:27124"
}
}
}
}
Or if running from source:
{
"mcpServers": {
"obsidian-todos": {
"command": "node",
"args": ["/path/to/obsidian-todos-api/mcp-server/dist/index.js"],
"env": {
"OBSIDIAN_API_URL": "http://localhost:27124"
}
}
}
}
Available Tools
list_todos
Lists all incomplete todos from your Obsidian vault.
Example:
// No parameters needed
await use_mcp_tool({
server_name: "obsidian-todos",
tool_name: "list_todos",
arguments: {}
});
add_todo
Adds a new todo to today's daily note.
Parameters:
text(string, required) - The todo text
Example:
await use_mcp_tool({
server_name: "obsidian-todos",
tool_name: "add_todo",
arguments: {
text: "Review pull requests"
}
});
update_todo
Updates an existing todo (mark complete, change text, etc.).
Parameters:
file(string, required) - Path to file containing the todoline(number, required) - Line number of the todo (0-indexed)text(string, optional) - New text for the todocompleted(boolean, optional) - Mark as complete/incomplete
Example:
await use_mcp_tool({
server_name: "obsidian-todos",
tool_name: "update_todo",
arguments: {
file: "Daily/2025-10-31.md",
line: 5,
completed: true
}
});
get_todo_stats
Get statistics about your todos.
Example:
await use_mcp_tool({
server_name: "obsidian-todos",
tool_name: "get_todo_stats",
arguments: {}
});
Development
# Watch mode for development
npm run dev
# Build for production
npm run build
# Test the server
echo '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' | node dist/index.js
Troubleshooting
Connection Issues
- Ensure Obsidian Local REST API is running
- Check the API URL is correct (default:
http://localhost:27124) - Verify the Todos REST API plugin is enabled in Obsidian
Authentication Errors
If you have authentication enabled in Local REST API:
- Set the
OBSIDIAN_API_KEYenvironment variable - Or update your Claude Desktop config with the API key
License
MIT
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。