What backup is recommended before using Obsidian MCP Server?
The README strongly advises backing up your Obsidian vault prior to use, recommending git or any backup method, because the server has read and write access and is in active development.
@StevenStavrakis
基本情報
A simple MCP server for Obsidian
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"obsidian": {
"command": "npx",
"args": [
"-y",
"obsidian-mcp",
"/path/to/your/vault"
]
}
}
}Read the contents of a note
Create a new note
Edit an existing note
Delete a note
Move a note to a different location
Create a new directory
Search notes in the vault
Add tags to a note
Remove tags from a note
Rename a tag across all notes
List and organize tags
List all available vaults (helps with multi-vault setups)
An MCP (Model Context Protocol) server that enables AI assistants to interact with Obsidian vaults, providing tools for reading, creating, editing and managing notes and tags. It is designed for Obsidian users who want to let AI assistants directly manage their note-taking vault.
Install manually by adding the server to your Claude Desktop configuration file, specifying the absolute path(s) to your Obsidian vault(s). The server runs via npx obsidian-mcp <vault-path>. A Smithery installation option is available but not recommended by the author. Requires Node.js 20 or higher.
The README strongly advises backing up your Obsidian vault prior to use, recommending git or any backup method, because the server has read and write access and is in active development.
Node.js 20 or higher and an Obsidian vault are required.
The server reads and writes directly to your local Obsidian vault directory. No cloud storage is mentioned; all data stays on your machine.
The tools have been tested but not thoroughly, and the server is in active development.
The server is configured to run as a local MCP server via stdio (command-line interface). No network transports are mentioned.

Model Context Protocol Servers

Local-first MCP server that snapshots your project's working state into one context object. Open-core, privacy-first.
Memlord is a remote MCP server available at https://app.memlord.com.
Groundwork: company memory for your AI tools. Docs public. Source closed. Public proof needs no account.
Shared, serverless, conflict-free memory for AI agents: a multi-machine drop-in for the MCP memory knowledge graph.
コメント