Skip to content

Commit cdd03f8

Browse files
authored
docs: Update README for v5.5 manual architecture
1 parent d2caf6a commit cdd03f8

File tree

1 file changed

+15
-28
lines changed

1 file changed

+15
-28
lines changed

README.md

Lines changed: 15 additions & 28 deletions
Original file line numberDiff line numberDiff line change
@@ -1,36 +1,30 @@
1-
# LuminOS Build Scripts
2-
19.11.2025
1+
# LuminOS Build Scripts
32

43
[![Build Status](https://img.shields.io/badge/build-passing-brightgreen)]()
54

65
Welcome to the official build repository for **LuminOS**.
76

8-
**LuminOS** is a lightweight, private, and intelligent Linux distribution based on Debian "Trixie". It features native integration of **Lumin**, a local, offline, and ethical AI assistant. By running this script, you will obtain the first version of LuminOS in the form of an .ISO file. Please note that this version of LuminOS is an alpha version and may not be stable and/or not contain all the features mentioned.
7+
**LuminOS** is a lightweight, private, and intelligent Linux distribution based on Debian 13 "Trixie". It features native integration of **Lumin**, a local, offline, and ethical AI assistant.
98

109
---
1110

1211
## 🏗️ Build Architecture (v0.2)
1312

14-
We utilize a **transparent, manual build process** (bypassing the complexity of `live-build` wrappers) to ensure maximum stability and customization.
13+
We choose a **transparent, manual build process** to ensure stability.
1514

1615
The master script `build.sh` orchestrates the entire pipeline:
17-
1. **AI Preparation:** Downloads Ollama and the Llama 3 model on the host to avoid chroot network issues.
18-
2. **Bootstrap:** Creates a pristine Debian base system using `debootstrap`.
19-
3. **Injection:** Copies scripts, assets, and the pre-downloaded AI models into the system.
20-
4. **Customization (Chroot):** Installs the Kernel, KDE Plasma, LuminOS Theme, and configures Lumin (the AI).
21-
5. **Assembly:** Compresses the filesystem (`SquashFS`) and generates a Hybrid ISO (BIOS/UEFI) using `grub-mkrescue`.
16+
1. **Bootstrap:** Creates a base Debian system using `debootstrap`.
17+
2. **AI Prep:** Pre-downloads Ollama and models on the host to avoid network issues
18+
3. **Injection:** Copies scripts and assets into the system.
19+
4. **Customization:** Executes scripts `02` through `07` inside the system (chroot) to install KDE, themes, and configure settings.
20+
5. **Assembly:** Compresses the filesystem and generates the ISO.
2221

2322
## 🚀 Prerequisites
2423

25-
To build LuminOS, you need a host machine running **Ubuntu 24.04 LTS** or **Debian 12+**.
26-
27-
* **Disk Space:** ~30 GB free.
28-
* **RAM:** 8 GB minimum recommended.
29-
* **Internet:** Required to download packages and the AI model (~5 GB).
30-
* **Privileges:** `sudo` access is required.
31-
32-
The script automatically installs the necessary dependencies:
33-
`debootstrap`, `squashfs-tools`, `xorriso`, `grub-pc-bin`, `grub-efi-amd64-bin`, `mtools`, `curl`, `rsync`.
24+
* Host: **Ubuntu 24.04 LTS** or **Debian 12+**.
25+
* Disk Space: ~30 GB free.
26+
* RAM: 8 GB minimum.
27+
* `sudo` privileges.
3428

3529
## 🛠️ Build Instructions
3630

@@ -46,18 +40,11 @@ The script automatically installs the necessary dependencies:
4640
```
4741

4842
3. **Retrieve the ISO:**
49-
Once completed (30 to 60 minutes depending on connection), the final image will be located here:
50-
`LuminOS-0.2-amd64.iso`
43+
The final image `LuminOS-0.2-amd64.iso` will be generated in the project folder (called build-scripts probably)
5144

5245
## 🤖 About Lumin (AI)
5346

54-
LuminOS v0.2 features a 100% local AI integration.
55-
* **Model:** Based on Llama 3 (may change to a newer version in futur releases)
56-
* **Privacy:** No data ever leaves your machine.
57-
* **Initialization:** The model is prepared during the build process. On the very first boot, a system service finalizes the Lumin setup in the background.
58-
* Lumin will be at the heart of LuminOS and will be integrated more deeply as future versions of LuminOS are released
47+
LuminOS v0.2 integrates a local AI based on Llama 3. The model is prepared during the build. On the first boot of the installed OS, a background service finalizes the setup automatically.
5948

6049
---
61-
62-
**License:** GPL-3.0
63-
*Built with passion for digital freedom. <3*
50+
**License:** GPL-3.0 :)

0 commit comments

Comments
 (0)