The simplest approach to structured Markdown is extracting frontmatter with regex and parsing the YAML yourself.

The manual approach [#the-manual-approach]

```typescript
const match = content.match(/^---\n([\s\S]*?)\n---\n/);
const frontmatter = match ? yaml.parse(match[1]) : {};
```

This works for simple cases, but it quickly breaks down:

* No body parsing — you still have unstructured text
* No typing — frontmatter is `any`
* No validation — typos in YAML are silent
* No positions — you can't point to where an error occurred

The datamark approach [#the-datamark-approach]

```typescript
import { parse } from "datamark";

const doc = parse(content);
// doc.frontmatter is parsed YAML
// doc.root is a structured section tree
```

For typed, validated data, use the Format SDK:

```typescript
import { datamark, inlineText } from "datamark";
import * as z from "zod";

const MyFormat = datamark({
  schema: z.object({ title: z.string() }),
  parse(doc) {
    const h1 = doc.root.children.find(n => n.type === "section") as any;
    const title = h1 ? inlineText(h1.heading.children) : "";
    return { title };
  },
});
```

| Feature                | Raw Frontmatter | datamark               |
| ---------------------- | --------------- | ---------------------- |
| Frontmatter extraction | ✅ Manual regex  | ✅ Built-in             |
| YAML parsing           | ✅ External lib  | ✅ Built-in             |
| Typed frontmatter      | ❌               | ✅ With Standard Schema |
| Body AST               | ❌               | ✅                      |
| Source positions       | ❌               | ✅                      |
| Validation errors      | ❌               | ✅ Structured           |

When to use raw frontmatter [#when-to-use-raw-frontmatter]

If you literally only need one key from the frontmatter and nothing else, raw extraction is fine. For anything more complex, datamark gives you structure, types, and error handling for free.