The Glob Loader

`glob()` pulls every matching file into the collection. Each file becomes one entry; its frontmatter is the entry's `data`.

3 min read Level 2/5 #astro#glob#loader

What you'll learn

Use `glob` with `pattern` and `base`
Understand the resulting entry id
Filter and nest files

The glob loader matches a set of files and turns each into an entry. It’s the right loader for blog posts, docs pages, recipes — anything where one file = one entry.

A Basic Glob

import { defineCollection, z } from "astro:content";
import { glob } from "astro/loaders";

const blog = defineCollection({
  loader: glob({ pattern: "**/*.md", base: "./src/blog" }),
  schema: z.object({ title: z.string() }),
});

pattern — globs to match (relative to base)
base — the folder to search from

Allowed File Types

.md, .mdx, .markdown, .json, .yaml, .yml — Astro picks the right parser per extension. For a single-file source (e.g. an authors list in one JSON), use the file loader instead (next lesson).

The Entry id

For each matched file, the id is the file path relative to base, without the extension.

base: ./src/blog
pattern: **/*.md

Files:
  hello-world.md            → id: "hello-world"
  advanced/effects.md       → id: "advanced/effects"

You’ll use this id later for routing and references.

Multiple Patterns

loader: glob({
  pattern: ["**/*.md", "**/*.mdx"],
  base: "./src/blog",
}),

Excluding Files

Lead any pattern with ! to exclude:

loader: glob({
  pattern: ["**/*.md", "!**/_*.md"],   // skip drafts that start with _
  base: "./src/blog",
}),

Up Next

When all your data is in one file — JSON or YAML.

The File Loader →