obsidian-mcp
MCP server that gives AI assistants access to an Obsidian vault, enabling folder navigation and note creation, reading, editing, moving, renaming, and deletion.
README
Obsidian MCP
An MCP (Model Context Protocol) server that gives AI assistants (Claude, Cursor, etc.) access to your Obsidian vault. Navigate folders, read notes, create, edit, move, rename and delete them — all from your AI chat.
Requirements
- Python 3.10+
- uv (recommended) or pip
- An Obsidian vault on your local machine
Installation
# Clone the repo
git clone https://github.com/your-username/obsidian-mcp
cd obsidian-mcp
# Copy the env template and set your vault path
cp .env.example .env
Edit .env and set your vault path:
OBSIDIAN_VAULT_PATH="/Users/you/Documents/MyVault"
To find your vault path in Obsidian: Settings → About → Vault path.
Running the server
Development (with MCP Inspector UI)
uv run mcp dev server.py
This opens the MCP Inspector in your browser where you can test all tools interactively.
Production
uv run server.py
Connecting to Cursor
Add this to your .cursor/mcp.json (or Cursor global MCP settings):
{
"mcpServers": {
"obsidian": {
"command": "python",
"args": ["/absolute/path/to/obsidian-mcp/server.py"],
"env": {
"OBSIDIAN_VAULT_PATH": "/path/to/your/vault"
}
}
}
}
Replace /absolute/path/to/obsidian-mcp/server.py with the actual path to server.py on your machine, and /path/to/your/vault with your vault path (find it in Obsidian: Settings → About → Vault path).
Connecting to Claude Desktop
Add this to your claude_desktop_config.json:
{
"mcpServers": {
"obsidian": {
"command": "uv",
"args": ["run", "/absolute/path/to/obsidian-mcp/server.py"],
"env": {
"OBSIDIAN_VAULT_PATH": "/path/to/your/vault"
}
}
}
}
Available Tools
| Tool | Description |
|---|---|
list_folder(folder_path) |
List subfolders and notes inside a vault folder. Leave empty for vault root. |
read_note(note_path) |
Read the full content of a note (path relative to vault root). |
create_note(note_path, content) |
Create a new note. Creates intermediate folders if needed. |
update_note(note_path, content) |
Replace the full content of an existing note. |
delete_note(note_path) |
Delete a note from the vault. |
rename_note(note_path, new_name) |
Rename a note keeping it in the same folder. |
move_note(note_path, destination_path) |
Move a note to a different folder. Can also rename as part of the move. Creates destination folders if needed. |
delete_folder(folder_path, recursive) |
Delete a folder. Use recursive=True to delete folders that contain files. |
All paths are relative to the vault root, e.g. "Projects/App/roadmap.md".
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。