opensite-ai/opensite-skills / linux-distro-selector
linux-distro-selector
VerifiedChoose the right Linux distribution for a developer or AI/ML workstation and confirm the hardware will work before installing. Use when someone asks which Linux distro to pick; is comparing Ubuntu, Fedora, Arch, NixOS, Pop!_OS, Mint, or atomic/immutable distros; is switching from Windows or macOS and unsure what to install; needs a distro matched to a use case (maximum stability, bleeding-edge GPU/AI work, reproducibility, laptop battery life, or a server); or wants to check whether a GPU, Wi-Fi chipset, or laptop is Linux-compatible. Advisory only: it produces a recommendation, not install commands.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 7.8 KB |
| agents/openai.yaml | config | 402 B |
| examples/nvidia-laptop-stability.md | markdown | 4.0 KB |
| references/activation.md | markdown | 2.9 KB |
| references/atomic-vs-traditional.md | markdown | 4.8 KB |
| references/distro-matrix.md | markdown | 11.8 KB |
| references/hardware-compat-triage.md | markdown | 6.6 KB |
| templates/hardware-intake-worksheet.md | markdown | 2.7 KB |
references/atomic-vs-traditional.md
View on GitHub# Atomic vs Traditional, Rolling vs LTS
Facts below are **as of 2026-07 — verify**. This file explains the two axes that
matter most for a senior AI developer's distro choice: the **update model**
(rolling / point-release / LTS / atomic) and whether to run an **immutable/atomic**
base with containerized dev.
---
## The two axes
1. **Release cadence:** rolling ↔ point-release ↔ LTS. Controls how fresh the
kernel/drivers/toolchains are and how much maintenance the user signs up for.
2. **Mutability:** traditional mutable root ↔ atomic/immutable image. Controls how
easily a bad update or driver can break the host, and how reproducible the
machine is across a fleet.
These are largely independent — e.g. Fedora Silverblue is *atomic + point-release*,
Bluefin is *atomic + continuously-updated image*, NixOS is *atomic + declarative*.
---
## Rolling vs point-release vs LTS
| Model | Freshness | Stability | Maintenance | Best for | Examples |
|---|---|---|---|---|---|
| **LTS / long point-release** | Low–med (backports help) | Highest | Lowest | Production parity, servers, "must not change under me" | Ubuntu LTS, Debian, Leap, RHEL/Alma/Rocky |
| **Regular point-release** | Med–high | High | Low–med (upgrade every ~6–12 mo) | Devs wanting recent tools + tested releases | Fedora, Ubuntu interim, Mint |
| **Rolling** | Highest | Medium (mitigated by snapshots/testing) | Highest (read update news; occasional breakage) | Bleeding-edge AI/GPU, newest toolchains | Arch, CachyOS, Tumbleweed, NixOS-unstable |
| **Atomic (any cadence)** | Depends on base | Highest (rollback) | Very low | Reliability + fresh dev via containers | Silverblue, Bluefin/Aurora, Aeon, NixOS |
---
## What "atomic / immutable" means
A read-only OS root delivered as a **versioned image** (rpm-ostree / bootc / Nix).
The user layers or, better, **containerizes** everything else on top.
**Pros for a senior AI dev:**
- **Can't break the host.** Bad update or driver? Roll back to the previous image at
boot in seconds.
- **Zero config drift** across laptop/desktop/fleet — huge for reproducible environments.
- **Automatic, atomic updates** — no half-applied state.
- **Clean separation:** stable, boring host + bleeding-edge, disposable dev layers.
**Cons / friction:**
- Installing something into the base image often needs a **reboot** (rpm-ostree layering) —
you are expected *not* to.
- Non-container binaries, kernel modules, VPN clients, and some proprietary tooling can be awkward.
- Debugging is different; less Stack Overflow coverage; steeper mental model.
- NixOS's reproducibility is the strongest but has the highest learning cost.
---
## Container-native dev is what makes atomic work
- **`toolbx`** (Fedora) and **`distrobox`** (distro-agnostic; podman/docker/lilipod) create
mutable containers that share `$HOME`, GPU, USB, Wayland/X11, and audio with the host.
- Run **Ubuntu, Arch, or a CUDA image** as a container on an immutable Fedora/openSUSE host —
pin per-project toolchains, export binaries to `~/.local/bin`, keep the host pristine.
- **Homebrew on Linux** (default on Bluefin) covers CLI tooling; **DevContainers/VS Code**
integrate natively.
This is the biggest shift since 2024: you no longer need a rolling *OS* to get rolling
*tools*. `uv`, `mise`, `rustup`, containers, and Homebrew let an LTS or atomic host run
bleeding-edge dev stacks. Reserve a rolling OS for when you need the newest
**kernel/Mesa/GPU driver** on brand-new hardware.
---
## When a senior AI dev should choose atomic
- Runs **multiple machines** and wants them identical → **yes**, atomic.
- Wants **newest dev tools but a host that never breaks** → **Bluefin/Aurora** or
**Silverblue/Kinoite** + distrobox.
- Is **reproducibility-obsessed** or manages a fleet → **NixOS**.
- Needs **deep host-level customization, out-of-tree kernel modules, or exotic hardware
hacks** → prefer a **traditional** distro (Arch/Fedora/Ubuntu).
---
## The AI-dev dilemma and its three resolutions
You need recent CUDA/ROCm/PyTorch/kernels **and** stability. Three good resolutions,
ranked by how much OS-tinkering the user enjoys:
1. **Rolling with a safety net:** CachyOS or Tumbleweed (Btrfs + Snapper auto-rollback).
Newest drivers, snapshot to undo.
2. **Atomic + containers:** Bluefin/Silverblue host (stable, rollback) + CUDA/bleeding-edge
in distrobox.
3. **LTS + selective freshness:** Ubuntu 26.04 LTS (now ships CUDA/ROCm natively) or Fedora,
pulling recent toolchains via containers / `uv` / `mise` / Homebrew.
**Tie-breaker:** Btrfs + Snapper/Timeshift auto-snapshots (CachyOS, Tumbleweed, Fedora opt-in)
massively de-risk rolling releases — near-atomic undo without going fully immutable. And if
the user's servers run Ubuntu/Debian, an Ubuntu LTS workstation minimizes "works on my machine"
drift.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |