Obsidian

MCP server enabling AI agents to perform natural knowledge discovery and analysis across Obsidian vault

python

20

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "pmmvr-obsidian-api-mcp-server": {
      "command": "uvx",
      "args": [
        "--from",
        "obsidian-api-mcp-server>=1.0.1",
        "obsidian-api-mcp"
      ],
      "env": {
        "OBSIDIAN_API_KEY": "YOUR_API_KEY",
        "OBSIDIAN_API_URL": "https://localhost:27124"
      }
    }
  }
}

You can transform your Obsidian vault into a powerful knowledge base for AI agents by running an MCP server that exposes a Local REST API. This server enables agents to perform advanced discovery, filtering, and content retrieval across your notes, with flexible searches, content access, and vault structure exploration.

How to use

To use the Obsidian MCP Server, you run the MCP server locally and connect your MCP client to the provided endpoint. This setup lets agents search notes, fetch complete note content with metadata, and browse vault structure. You can chain together searches and content retrieval to build multi-step workflows such as filtering notes by path and date, extracting action items from meetings, or mapping notes by tags to support knowledge work.

How to install

Prerequisites include Python and a working local Obsidian vault with the Local REST API plugin installed and configured. You will also interact with an MCP client to connect to the server.

Configuration and usage details

The MCP client can be configured to connect to the Obsidian MCP Server via a local endpoint and an API key if you enable authentication. You will typically provide the API URL (default https://localhost:27124) and an API key if configured.

Additional setup and examples

Security considerations include keeping your API key out of version control and using environment variables or a .env file to manage secrets. For HTTP connections, you can point to http://localhost:27123 if you are not using TLS.

Troubleshooting tips

If you encounter SSL certificate issues with self-signed certificates, ensure the Obsidian API URL is reachable and that SSL verification is appropriately configured for your environment. If the server fails to start, verify that Python and the required dependencies are installed and that the Obsidian Local REST API plugin is running in your vault.

Notes on security and best practices

Only expose the MCP server to trusted clients. Use API keys when available, rotate keys periodically, and avoid embedding secrets in scripts. Use a .env file or system environment variables to manage credentials securely.

Tools and endpoints overview

The server provides three primary tools for agents to interact with your Obsidian vault: searching notes with flexible filters and full content retrieval, retrieving complete content and metadata for a specific note, and navigating the vault structure to understand folder and file organization.

What you can do with the MCP server

With the MCP server running, you can perform tasks such as locating notes by path and title, retrieving full note bodies and metadata for analysis, and exploring the vault layout to support planning and cross-reference workflows. This enables multi-step workflows like risk assessment from project notes, gap analysis across research notes, and action item extraction from meeting notes.

Developer notes

When you want to customize or extend the server, you can build from source, install development dependencies, and run the server locally for testing. The following sections provide concrete steps to install, configure, and run the MCP server in development or production modes.

Available tools

search_vault

Advanced search with flexible filters and full content retrieval across your vault notes. Supports text and regex queries, path and date filters, tags, and content inclusion options.

get_note_content

Retrieve complete content and metadata for a specific note by its path, enabling precise analysis and reference tracking.

browse_vault_structure

Navigate the vault directory structure efficiently, with options to include files and to browse recursively for fast overview.