Skip to content

opensite-ai/opensite-skills / linux-dev-workstation

linux-dev-workstation

Unverified

Configure a senior-grade Linux development environment on the machine: GPU drivers and compute stacks (NVIDIA CUDA, AMD ROCm, Intel XPU, Apple Silicon/Asahi), shell and terminal (zsh/fish, starship, modern CLI tools, tmux/zellij), advanced editor setup (Neovim/Vim, VS Code, Zed), language toolchains for Rust, Python, Ruby, and JavaScript/TypeScript via mise/uv/pixi, Docker/Podman containers, reproducible declarative setup (chezmoi, Ansible, Nix/home-manager), and security hardening (SSH keys, commit signing, secrets, firewall, backups). Use when setting up or upgrading a Linux dev box, installing GPU drivers, managing language versions and updates, configuring editors or dotfiles, or making a workstation reproducible. It runs commands on the machine.

Linux Dev Workstation

Configure a senior-grade Linux development environment directly on the machine — GPU drivers, shell, editors, language toolchains, containers, reproducibility, and security.

Mode: Configurator — this skill runs commands on the machine. Follow the SAFETY PROTOCOL below on every stage.

⚠️ SAFETY PROTOCOL (read first, applies to every step)

  1. Detect first. Never assume the distro, package manager, GPU vendor, or desktop. Run the detection step and branch from what you find.
  2. Make steps idempotent. Prefer commands that are safe to re-run (mise use, apt install, checking "already installed" before acting).
  3. Verify after each stage. Every stage below states a verify command. Run it and confirm success before moving on.
  4. Confirm before destructive/irreversible actions. Require an explicit user "yes" before: disk/partition ops, driver swaps or removals, blacklisting kernel modules, firewall enable/rule changes (risk of locking out SSH), chsh (default-shell change), and any rm. State exactly what will change and how to undo it.
  5. Never paste secrets into the shell history or logs. Use the secrets tools in security-hardening.md.
  6. Currency guardrail. Embedded version numbers are labeled "as of 2026-07 — verify." Verify current driver, toolchain, and package versions with a web search before installing.

Skill Resources

References (load on demand for copy-pasteable depth):

Templates:

Example:

Step 1 — Detection (always first)

Gather the facts that drive every branch. Do not edit anything yet.

# Distro + package manager
. /etc/os-release && echo "$PRETTY_NAME"
command -v apt dnf pacman zypper 2>/dev/null   # which package manager

# GPU vendor(s)
lspci | grep -Ei 'vga|3d|display'

# Desktop session type (Wayland vs X11) and form factor
echo "$XDG_SESSION_TYPE"
ls /sys/class/power_supply/ | grep -qi bat && echo "laptop" || echo "desktop"

# Kernel + Secure Boot state (affects out-of-tree driver signing)
uname -r
mokutil --sb-state 2>/dev/null || echo "Secure Boot state unknown"

Record: distro/package manager, GPU vendor(s), Wayland/X11, laptop/desktop, Secure Boot on/off. Every later stage branches on these.

Step 2 — Ordered bring-up

Run in order; each stage is idempotent and ends with a verify command. Confirm the verify passes before continuing.

#StageReferenceVerify command
1GPU drivers + computegpu-drivers.mdnvidia-smi / rocm-smi / xpu-smi discovery / glxinfo -B
2Shell + terminalshell-terminal.mdecho $SHELL; starship --version; rg --version
3Editorseditors.mdnvim --version; code --version; zed --version
4Language toolchainslanguage-toolchains.mdmise ls; rustc -V; uv --version; ruby -v; node -v
5Containerscontainers.mddocker run --rm hello-world or podman run --rm hello-world
6Reproducibilityreproducibility.mdchezmoi doctor; ansible --version
7Security hardeningsecurity-hardening.mdssh-add -l; sudo ufw status / firewall-cmd --state

Driver stage is the only one that is potentially destructive (module blacklist, DKMS, Secure Boot MOK enrollment, reboot). Treat it under the SAFETY PROTOCOL: confirm before acting, keep X11 as a fallback, verify with nvidia-smi before rebooting into a possibly-black-screen session.

After stage 7, hand off to the linux-ai-dev-stack skill for local LLM inference, AI coding agents, MCP servers, ML frameworks, and remote GPU. This skill only builds the base; the AI toolchain lives there and assumes the working driver stack you just installed.

Language toolchains quick map

Use mise as the meta version-manager for everything except Rust (rustup) and Python (let uv own Python + envs). Full depth, LSP, and linters are in language-toolchains.md. Versions as of 2026-07 — verify.

LanguageManagerInstallPer-project pinUpdate
Rustrustupcurl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | shrust-toolchain.toml or mise.tomlrustup update
Pythonuvcurl -LsSf https://astral.sh/uv/install.sh | shuv python pin 3.13 + uv.lockuv self update; uv lock --upgrade
Rubymisemise use ruby@3.4.tool-versions / mise.tomlmise upgrade ruby; bundle update
Node/TSmisemise use node@lts.tool-versions / mise.tomlmise upgrade node; pnpm up
(meta)misecurl https://mise.run | shmise.toml (see template)mise self-update; mise upgrade

Reproducibility decision (pick one backbone)

Capture the machine so the next one is one command. Detail + matrix in reproducibility.md.

  • Pragmatic senior, multiple machines → chezmoi + Ansible + mise/uv (imperative-repeatable). One-liner: sh -c "$(curl -fsLS https://get.chezmoi.io)" -- init --apply <github-user>.
  • Declarative but keep your distro → home-manager (flakes) + direnv. One-liner: install Determinate Nix, then home-manager switch --flake ..
  • Reproducibility/rollback purist → NixOS + home-manager + flakes (nixos-rebuild switch --flake .).
  • Team parity / onboarding / CI → devcontainers (see templates/devcontainer.json) plus chezmoi for the personal layer.

See also

  • linux-distro-selector — if the distro is not chosen yet.
  • linux-install-planner — if the OS is not installed yet (Windows→Linux).
  • linux-ai-dev-stack — the AI toolchain (local LLM, agents, MCP, remote GPU); run it after this skill's driver stage succeeds.
rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit.Stars: 464kUpdated Claude CodeCodexVerifiedInstall
rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure.Stars: 464kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 464kUpdated Claude CodeCodexVerifiedInstall
ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first.Stars: 192kUpdated Claude CodeCodexVerifiedInstall