MDV — Markdown Data & Visualization

Write documents, dashboards, and slides in a Markdown superset. Add charts, KPI cards, tables, and styled regions with nothing more complicated than fenced code blocks and named styles.

.mdv is strict CommonMark plus four additions:

1. YAML front-matter for title, theme, named styles, and dataset references.
2. Fenced blocks for data/visuals: ```chart type=bar x=region y=sales.
3. ::: containers for styled regions and layout: ::: callout / ::: columns.
4. ::: toc for an auto-generated table of contents.

No selectors, no classes, no expressions, no code. Themes provide defaults, named styles give reusable looks, the renderer does the rest.

Quick look

---
title: Q1 Report
theme: report
data:
  sales: ./data/sales.csv
---

::: toc
:::

# Q1 Results

```stat
label, value, delta
Total revenue, $2.06M, +14%
New customers, 1238, +8%
```

```chart type=line data=sales x=month y=revenue series=region yFormat=currency title="Monthly revenue"
```

Renders to self-contained HTML (charts are inline SVG, no JS runtime) and PDF. Lives inside VS Code via a side-by-side preview.

Getting started

git clone <repo> mdv
cd mdv
npm install
npm run build

# Render an example
node packages/mdv-cli/dist/index.js render examples/09-full-report.mdv

# Or: live preview with auto-reload
node packages/mdv-cli/dist/index.js preview examples/09-full-report.mdv

See docs/getting-started.md for a walkthrough.

Documentation

• Getting started — install, author your first file, see it rendered.
• Syntax reference — front-matter, fenced blocks, ::: containers.
• Charts & stats — every visualization type with all options.
• Data — inline CSV / JSON / file-referenced datasets.
• Themes & styles — built-in themes and how to define named styles.
• CLI — render, preview, export --pdf.
• VS Code extension — side-by-side live preview.
• Publishing the VS Code extension — Marketplace workflow.

Examples

examples/ contains 10 sample files covering every feature. Rendered HTML is in examples/out/.

Status