Markdown Task Extractor

Markdown Task Extractor MCP Connector for Claude

A+

Instantly scan any local folder of Markdown notes (Obsidian, Notion, Logseq) and aggregate every scattered '- [ ]' to-do item into a single, structured list for your AI.

1 tools Official Updated Jun 28, 2026 Official Vinkius Partner

If you use Obsidian, Logseq, or Notion, your tasks are probably scattered across dozens of different daily notes and project files. When you ask your AI, 'What are my pending tasks today?', it has no idea because it can't read your local vault effectively.

This MCP uses a hyper-fast glob pattern to scan hundreds of local .md files in milliseconds. It extracts every - [ ] (pending) and - [x] (completed) task, along with the specific file it came from, and feeds it directly into your AI chat context. It transforms your local vault into a centralized AI task dashboard.

The Superpowers

  • Vault-Wide Aggregation: Turns your scattered notes into a centralized task dashboard.
  • Zero Config: Just give the AI the absolute path to your notes folder.
  • Lightning Fast: Uses fast-glob to scan 1,000+ files in under 50ms.
  • Status Aware: Perfectly distinguishes between open and completed tasks.
task-trackingmarkdownglob-patterntodo-listnote-takingtask-aggregation

1 tools expose this connector's capabilities to your AI agent.

extract_markdown_todos

Provide the absolute directory path to scan. Scan a local directory of Markdown files (Obsidian, Notion, Logseq) and extract all open and completed tasks (- [ ] and - [x])

See how to talk to your AI agent using Markdown Task Extractor.

Scan my Obsidian vault at C:/Notes and list all my pending tasks grouped by file.

I found 12 pending tasks across 4 different files. Here is the grouped breakdown:

Look through my Notion exports folder and tell me how many tasks I completed this week.

Based on the `- [x]` markers, you have completed exactly 24 tasks.

Find all tasks in my project folder that contain the hashtag '#urgent'.

I found 3 open tasks marked with #urgent: 1. Fix database deployment (backend.md) 2. Call investor (meetings.md)

Yes! It uses an optimized `fast-glob` engine that can recursively scan thousands of nested folders and markdown files almost instantly without crashing.

Related Connectors