mcp-obsidian-tools
A Model Context Protocol (MCP) server that allows MCP clients (Cursor, VSCode, Claude Desktop, etc.) to read and search any directory containing Markdown notes, such as an Obsidian vault. This server exposes a rich, read-only toolkit of obsidian_-prefixed MCP tools for working with vault metadata (tags, links, frontmatter), filenames, and more
claude mcp add --transport stdio dp-veritas-mcp-obsidian-tools node /absolute/path/to/mcp-obsidian-tools/dist/index.js /path/to/your/vault
How to use
The Obsidian Model Context Protocol Tools MCP server provides read-only access to an Obsidian vault (or any directory containing Markdown notes) via MCP. It exposes a rich set of obsidian_ tools designed to help clients search, inspect, and analyze vault content without modifying files. Available tools include: obsidian_search_notes for filename-based searches, obsidian_read_notes to fetch note contents (with optional headersOnly for quick structure extraction), obsidian_list_tags to enumerate tags with counts, obsidian_notes_by_tag to find notes by tag, obsidian_get_frontmatter to retrieve YAML frontmatter, obsidian_backlinks to discover linking notes, obsidian_search_content for full-text content search, obsidian_query for natural-language vault queries, and obsidian_count_files to gauge vault size by folder. These tools enable comprehensive vault exploration, metadata extraction, and content discovery in a read-only, secure manner restricted to the configured vault path.
To use it from an MCP client (e.g., Cursor, VSCode, Claude Desktop), configure a server entry that points to the Node.js runtime and the built distribution of this tool, along with the absolute path to your vault. Once configured, invoke the obsidian_* tools with appropriate parameters as described by your client’s MCP interface. Path handling requires absolute paths for both the server runtime and the vault, ensuring the MCP server can safely access only the intended directory.
How to install
Prerequisites:
- Node.js v18 or higher
- npm (comes with Node.js)
- An Obsidian vault or directory containing Markdown files
Installation steps:
-
Clone the repository: git clone https://github.com/dp-veritas/mcp-obsidian-tools.git cd mcp-obsidian-tools
-
Install dependencies: npm install
-
Build the project: npm run build
This compiles the TypeScript source and outputs the executable distribution files to the dist/ directory.
-
Run the server locally (example): node dist/index.js path/to/your/vault
Note: The exact invocation may vary depending on how you deploy; use the mcp_config guidance to point to the built dist/index.js and your vault.
Additional notes
Tips and considerations:
- Ensure Node.js v18+ is installed and accessible in PATH.
- Use absolute paths for both the server executable (dist/index.js) and the vault directory in MCP client configurations.
- The server is read-only and restricts access to the vault directory (and its real/symlinked path) for security.
- If you encounter issues where tools don’t appear in your MCP client, restart the client after updating configurations and verify vault path accessibility.
- If you run globally (npm install -g .), configure the server with command "mcp-obsidian-tools" and provide the vault path as an argument per the global configuration example.
- For large vaults, consider using obsidian_count_files to understand vault size and structure before running content-wide searches.
Related MCP Servers
mcp-for-beginners
This open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable, and secure AI workflows from session setup to service orchestration.
mcp-python-interpreter
MCP Python Interpreter: run python code. Python-mcp-server, mcp-python-server, Code Executor
mcp-zen
Enhanced Zen MCP Server with 'zen' default tool and improvements
anki
MCP server for anki
openai -agent-dotnet
Sample to create an AI Agent using OpenAI models with any MCP server running on Azure Container Apps
unity-editor
An MCP server and client for LLMs to interact with Unity Projects