diff --git a/docs/_config.yml b/docs/_config.yml
new file mode 100644
index 0000000..55d849d
--- /dev/null
+++ b/docs/_config.yml
@@ -0,0 +1,4 @@
+theme: jekyll-theme-cayman
+title: ChimpStackr
+description: Open-source focus stacking for Windows, macOS, and Linux
+show_downloads: true
diff --git a/docs/algorithms.md b/docs/algorithms.md
index aa9fa4d..0ef0d0b 100644
--- a/docs/algorithms.md
+++ b/docs/algorithms.md
@@ -1,5 +1,13 @@
+---
+layout: default
+title: Stacking Algorithms
+description: How each of ChimpStackr's four fusion methods works
+---
+
 # Stacking Algorithms
 
+[← Back to home](index.html)
+
 ChimpStackr implements four image fusion methods. This document explains how each works and when to use it.
 
 ## Laplacian Pyramid (default)
diff --git a/docs/index.md b/docs/index.md
new file mode 100644
index 0000000..c0279fc
--- /dev/null
+++ b/docs/index.md
@@ -0,0 +1,73 @@
+---
+layout: default
+title: ChimpStackr
+description: Open-source focus stacking for Windows, macOS, and Linux
+---
+
+**ChimpStackr** is a free, open-source focus stacking application for Windows, macOS, and Linux. It merges a series of differently-focused photos into a single, fully-sharp image — with a full GUI and a headless CLI.
+
+[Download](https://github.com/noah-peeters/ChimpStackr/releases){: .btn } [View on GitHub](https://github.com/noah-peeters/ChimpStackr){: .btn }
+
+![ChimpStackr main window](screenshots/main.png)
+
+## Features
+
+- **4 stacking algorithms** — Laplacian Pyramid, Weighted Average, Depth Map, Exposure Fusion (HDR)
+- **Automatic alignment** — translation-only or rotation + scale correction (focus breathing)
+- **16-bit pipeline** — full bit-depth preservation from RAW to output
+- **Auto-crop** — removes black edges from alignment shifts
+- **Auto-tuning** — parameters auto-detected from image resolution
+- **GUI + CLI** — full graphical interface and headless command-line tool
+- **Cross-platform** — native builds for Windows, macOS, Linux
+- **Pause / resume / cancel** — control long-running stacks
+- **Before/after comparison** — slider viewer for input vs output
+- **Drag & drop** — drop image files or folders directly into the app
+
+![Processing view](screenshots/processing.png)
+
+## Download
+
+Pre-built packages are on the [Releases](https://github.com/noah-peeters/ChimpStackr/releases) page:
+
+| Platform | Download | Notes |
+|---|---|---|
+| **Windows** | `ChimpStackr-Windows.zip` | Extract and run `chimpstackr.exe` |
+| **macOS** | `ChimpStackr-macOS.dmg` | Open DMG, drag to Applications |
+| **Linux** | `ChimpStackr-Linux-x86_64.AppImage` | `chmod +x` and run |
+
+## CLI quickstart
+
+Headless focus stacking, no GUI required:
+
+```bash
+# Basic stack
+chimpstackr-cli --input images/*.jpg --output result.tif
+
+# Align + stack with auto parameters
+chimpstackr-cli -i images/*.jpg -o result.tif --align --auto --auto-crop
+```
+
+Available methods: `laplacian` (default), `weighted_average`, `depth_map`.
+
+## Documentation
+
+- [Stacking Algorithms](algorithms.html) — how each of the four methods works and when to use it
+- [Packaging & Distribution](packaging.html) — building native packages for each platform
+
+## Build from source
+
+Requires Python 3.9–3.13.
+
+```bash
+git clone https://github.com/noah-peeters/ChimpStackr.git
+cd ChimpStackr
+python -m venv .venv
+source .venv/bin/activate   # .venv\Scripts\activate on Windows
+pip install -r requirements.txt
+python src/run.py           # GUI
+python -m src.cli --help    # CLI
+```
+
+## License
+
+[GPL-3.0](https://github.com/noah-peeters/ChimpStackr/blob/master/LICENSE).
diff --git a/docs/packaging.md b/docs/packaging.md
index 94e75f0..648f47c 100644
--- a/docs/packaging.md
+++ b/docs/packaging.md
@@ -1,5 +1,13 @@
+---
+layout: default
+title: Packaging & Distribution
+description: Building native ChimpStackr packages for each platform
+---
+
 # Packaging & Distribution
 
+[← Back to home](index.html)
+
 ChimpStackr uses PyInstaller to create native packages for all platforms. Builds are automated via GitHub Actions but can also be done locally.
 
 ## Quick Reference