Obsidian Zettelkasten MCP Server
Enables AI models to read, search, and write notes in an Obsidian vault following the Zettelkasten method, with tools for semantic search, memo reading/writing, and dialogue saving.
README
Obsidian Zettelkasten MCP Server
This is an MCP (Model Context Protocol) server designed to connect your Obsidian Vault with AI models (like Claude in Cursor).
Intended Use Case
This tool is specifically designed for users who manage their knowledge with the following setup:
- Zettelkasten Method: Organizing notes using a directory structure like
000_Slipboxin an Obsidian Vault. - AI Dialogue Aggregation: Consolidating AI conversations into specific directories like
000_Slipbox/ai_dialogues(or11_claude_dialogues, etc.) to track the evolution of your thoughts. - Git Submodule Workflow: Managing the Obsidian Vault as a Git submodule within a project workspace to treat code and knowledge as an integrated system.
Tools Provided
search_memos: Semantic search across your Vault. Finds relevant notes even if keywords don't match exactly.read_memo: Read the content of a specific note.write_memo: Save a plain Markdown file to a specific path (for drafts or documents).write_dialogue: Save a conversation as a formatted dialogue note (with date and provider).list_recent_memos: List recently updated notes.
Setup Instructions
1. Install Dependencies
pip install -r requirements.txt
2. Register in Cursor
You can enable this server in Cursor settings to allow Claude to read and write your notes directly.
- Open Cursor Settings (Cmd+,).
- Navigate to Features > MCP.
- Click + Add New MCP Server.
- Enter the following information and click Save:
- Name:
ObsidianMemo - Type:
command - Command:
/path/to/your/venv/bin/python /path/to/server.py
For more advanced configuration (e.g., in mcp.json), use the following structure:
{
"mcpServers": {
"ObsidianMemo": {
"command": "/path/to/your/venv/bin/python",
"args": [
"/path/to/server.py"
],
"env": {
"OBSIDIAN_VAULT_PATH": "/path/to/your/obsidian/vault"
}
}
}
}
Usage Examples
Once registered, you can prompt Claude like this:
- "Search for notes about 'LLM architecture' from last month."
- "Save this discussion as a new note titled 'MCP Integration Ideas' in the ai_dialogues folder."
- "What are the 5 most recently updated notes?"
Important Notes
- This server communicates via
stdio(standard input/output). - Note saving defaults to
000_Slipbox/ai_dialogues/, but you can specify existing folders like11_claude_dialogues. - Set the root path of your Vault by either editing
DEFAULT_VAULT_PATHinserver.pyor setting theOBSIDIAN_VAULT_PATHenvironment variable.
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。