home / mcp / markdown frontmatter mcp server

Markdown Frontmatter MCP Server

Provides a queryable view of Obsidian-style Markdown notes by front matter to surface recent thinking and relevant topics.

Installation
Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "caffeinatedwes-markdown-frontmatter-mcp": {
      "url": "https://mcp.example.com/markdown-frontmatter",
      "headers": {
        "KB_PATH": "/path/to/your/obsidian/vault"
      }
    }
  }
}

You run a lightweight MCP server that queries Markdown files by front matter to surface notes based on tags and recency. It enables you to ask an AI for things like what you’ve been thinking about a topic lately, using your own Markdown knowledge base as the data source.

How to use

You connect an MCP client to the server and start issuing the query_recent_notes tool to retrieve notes filtered by created/updated dates and tags. The server reads Markdown files, extracts YAML front matter, and returns structured results such as file path, title, tags, created/updated dates, and an excerpt. Use this to build AI prompts like: “What have I been thinking about [X] lately?” or “Catch me up on what’s been on my mind.” When you specify tags and a time window, you get relevant, recent notes that match your filters.

What you can accomplish with the tool

- Filter notes by tags to focus on specific topics.

- Narrow results to a date window using created or updated dates.

- Limit results and search within designated folders.

- Get concise previews with the note excerpt and metadata to decide relevance at a glance.

Available tools

query_recent_notes

Queries Markdown files by YAML front matter to return notes matching tags and a date window. Returns file path, title, tags, created/updated dates, and an excerpt.