- LIST
- GROUP
- QUERY
- HUMAN
Short answer
How can I find all my draft notes using AI?
Your AI uses the harvester to read the YAML frontmatter of your markdown files. It identifies files with specific status tags, like 'draft' or 'pending', so you can instantly pull a list of unfinished work. You'll see exactly which files need your attention next.
Where your notes land.
The results of your query.
The AI organizes your files based on the metadata you've already written.
LIST
A filtered file list
The AI identifies every file matching your status criteria. You get a clean list of paths to your unfinished notes.
GROUP
Categorized note sets
The AI groups notes by their specific status tags. This lets you see all 'pending' notes separately from 'draft' notes.
QUERY
Targeted content searches
The AI finds notes that meet multiple criteria at once. You can ask for notes that are both 'draft' and 'urgent'.
HUMAN
Manual review required
The AI only reads the metadata. You still need to open the files to actually write or edit the content.
The workflow
What your AI does when <the thing arrives>.
The AI handles the heavy lifting of scanning your local directory and parsing the metadata.
Scan directory
The AI looks through your specified markdown folder to find all available files.
harvest_markdown_frontmatterParse frontmatter
The AI extracts the YAML blocks from the top of each markdown file.
harvest_markdown_frontmatterFilter by status
The AI matches the extracted tags against your specific request, like finding only 'draft' files.
harvest_markdown_frontmatterPresent results
The AI gives you a list of the files that matched your filter so you can act on them.
harvest_markdown_frontmatter
Try it
Copy these to start.
Use these prompts to begin interacting with your notes.
Starting points
Swap out the status names like 'draft' or 'pending' with whatever tags you actually use in your files.
Markdown Frontmatter Harvester Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_pSzsBgFkLuKZKUA1Ej60Dr18di9s0JV9d52i5cdI/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Markdown Frontmatter Harvester capabilities are ready to use.
{
"mcpServers": {
"markdown-frontmatter-harvester-mcp": {
"url": "https://edge.vinkius.com/vk_preview_pSzsBgFkLuKZKUA1Ej60Dr18di9s0JV9d52i5cdI/mcp"
}
}
}Find all my markdown notes where status is draft.
List all files in my notes folder marked as pending.
Show me all notes with the tag 'todo' in the frontmatter.
Which notes are currently marked as 'in-progress'?
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Markdown Frontmatter Harvester once, then ask.
Just click to connect. Your credentials stay encrypted, and you can immediately start asking your AI about your local files.
Connect Markdown Frontmatter Harvester to your AIFAQ
How this task behaves.
- 01
Can the AI edit my notes?
No. This tool only reads the frontmatter to find files. It does not change your note content or your tags.
- 02
Does the AI delete any files?
No. The AI has no capability to delete, move, or rename your markdown files.
- 03
Can it find notes without frontmatter?
No. The tool specifically looks for YAML frontmatter. If a note doesn't have a status tag in the frontmatter, the AI won't find it using this method.
- 04
Does it work with Obsidian or Logseq?
Yes, as long as your notes are stored as local markdown files with standard YAML frontmatter.
- 05
Can I filter by multiple tags at once?
Yes. You can ask the AI to find notes that have both a 'draft' status and a specific 'project' tag.
More questions about Markdown Frontmatter Harvester? The Connector page answers them. See everything the Markdown Frontmatter Harvester Connector can do
Connect Markdown Frontmatter Harvester to Claude, Cursor, ChatGPT & more
