DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to draw an org chart or service map with your own types with Md2 HD

Define types and an inverse block in your own files to color and lay out a custom map.

  • Research
  • Intermediate
  • 30 minutes
  • From the README
  • Any agent

Who it is forA team lead or operations owner who wants an org chart or service dependency map built from notes they already keep.

The problemGeneric diagram tools need separate data entry, and the notes describing the structure drift out of date.

The outcomeA map of your chosen types, with each relationship reading in the voice of its node, such as employs and works at.

What you need first

  • Node.js and npm installed, so that npx works
  • A markdown file or a folder of markdown notes
  • A modern web browser

Steps

  1. Add a type: map block at the top of a markdown file, with a title and an inverse entry for each relationship pair.
  2. Give each node a frontmatter block with an id, a type, a title, and a rel: list.
  3. Run npx md2hd notes/ to open the map in the browser.
  4. Click the Code toggle in a node's pane and edit the source file to watch the map redraw.

Prompt for your agent

Using the md2hd repo at evan-steinhilb/md2hd, write a type: map block and node frontmatter for the org chart described in <placeholder>. Expected output: markdown files with an inverse entry such as works_at: employs. Verify by running `npx md2hd <placeholder>` and checking that the relationship labels read correctly on both nodes.

Caveat. Md2 HD reads your own type names and labels from the files, so the map is only as clear as the types and inverse pairs you write.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for evan-steinhilb/md2hd and "Draw an org chart or service map with your own types"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer