agentleFS
johnplantadajohnplantada/claude-skills/skills/runtime-versions/SKILL.md

Runtime Versions

Skill0 starsChanged 4 months agoMITView on GitHub

Find, set up, repair, upgrade, and optimize your language runtimes (node, python, go, ruby…) with ONE tool — mise — instead of a pile of nvm/asdf/pyenv/homebrew-node managers that fight over PATH. Use to FIND which runtime/version/scope to install (search mise's versions, LTS vs latest, global vs per-project), SET UP mise (greenfield, or migrate off the sprawl — capture versions, activate, neutralize old managers), REPAIR a broken resolution (`node` differs per shell, an old shim wins, mise not activated), UPGRADE versions day-to-day (`mise use`, bump, prune, honor `.tool-versions`/`.nvmrc`), or OPTIMIZE a working setup (residual sprawl, leftover shim dirs, global versions lagging intent). Every change is verified in a clean environment, in both shells, before old managers are removed.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read ~/.config/devenv/config.toml before prompting.

    line 83

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Runtime versions

Manage language runtimes with one manager — mise — and end the nvm/asdf/pyenv/rbenv/fnm/homebrew-node sprawl where every tool has its own shims, its own PATH entry, and its own opinion about which node you get. mise is a single fast shim layer that reads the .tool-versions/.nvmrc/mise.toml files the others left behind, so migration is additive.

Route to a workflow

The user wants to…Workflow
Choose which runtime / version / scope to install (before installing)reference/find.md
Stand up mise — greenfield, or migrate off nvm/asdf/pyenv/… safelyreference/setup.md
Fix a broken resolution — node differs per shell, an old shim wins, mise not activereference/repair.md
Set / install / bump versions day-to-day, global or per-projectreference/upgrade.md
Review a working setup — residual sprawl, leftover dirs, global lags intentreference/optimize.md

The scripts (prefer these over ad-hoc bash)

Stable, tested, read-only helpers live in scripts/ — call them instead of re-composing the command -v / which -a / mise which / env -i blocks each session. Run by absolute path from this skill's directory. They handle a missing manager/shell/mise gracefully.

NeedScript
Choose what to install (find): is a tool mise-managed, recent versions + @lts, current owner, brew/legacy flagscripts/runtime_find.py [tool …]
Sprawl hunt: which managers are present, their rc hooks + versions, brew runtimes, shim dirs (run first for setup/repair/optimize)scripts/detect_managers.py
Who resolves a runtime (current shell): command -v, real path, owning manager, mise's opinion, dupesscripts/tool_resolve.py [tool …]
The real per-shell test — clean-env (env -i) login resolution in zsh/fish + mise-shims countscripts/shell_resolve.py [zsh|fish|both] [tool …]
mise health + inventory: activated? shims dir, problems, current, lsscripts/mise_status.py

Full contract + a worked example: scripts/README.md. The inline bash snippets in this file and the reference .mds show what the scripts run under the hood — reach for them for novel one-off checks, not to re-type the routine diagnostics.

Discovery (always run first)

Run scripts/detect_managers.py (which managers, rc hooks, brew runtimes, shim dirs), then scripts/tool_resolve.py (what each tool resolves to now) and, for the per-shell truth, scripts/shell_resolve.py both. Under the hood, that is:

mise --version 2>/dev/null || echo "mise NOT installed"
command -v mise nvm asdf pyenv rbenv fnm                  # which managers are present
brew list --formula 2>/dev/null | grep -E '^(node|python|ruby|go)(@|$)'  # homebrew runtimes
mise which node python go 2>/dev/null                     # what mise would resolve (if active)
command -v node python go                                 # what THIS shell actually resolves
  • Multiple managers present, or a runtime installed via both brew and a manager, is the sprawl this skill fixes — different shells silently get different versions. Review it in optimize.md; an active per-shell divergence is repair.md.
  • Shells activate these managers, so the truth is per-shell. Always check zsh and fish (scripts/shell_resolve.py both).

Core principles

  1. mise-first, additive migration. mise reads existing .tool-versions/.nvmrc, so adopting it doesn't break projects. Set up mise, verify it, then neutralize the old managers.
  2. Verify in a CLEAN environment. A shell spawned inside your current session inherits a polluted PATH. Prove resolution with scripts/shell_resolve.py both (clean env -i in each shell) — see verification.md.
  3. Don't remove old managers until mise is verified working in both zsh and fish. Back up shell config first; shell edits are owned by the shell-sync skill (below), not reimplemented here.
  4. Don't assume state. Re-run discovery each time — which manager wins depends on shim PATH order.

Cross-skill coordination (reference, don't reimplement)

  • shell-sync — mise activation adds shims to PATH; the canonical shell owns the mise activate line and the mirror is regenerated. Route all shell-config edits through it.
  • dotfiles — track ~/.config/mise/config.toml via chezmoi so global versions are reproducible.
  • brew-doctor — install mise via brew; uninstall nvm/asdf/node@x after migration verifies.
  • nvim-config — LSP servers run on the mise-managed runtime; a version change can move them.

Settings

Read ~/.config/devenv/config.toml before prompting; if a [runtime-versions] section exists, use it as defaults (precedence: explicit answer this session > config.toml > ask). Had to ask? Offer to save the answer back. Managed by the devenv skill. Keys honored:

  • manager — the version manager to standardize on (e.g. "mise").
  • tools — map of global runtime versions, e.g. node = "lts", python = "3.13", go = "latest".

Safety

  • Switching managers changes which runtime a project resolves — verify per-project .tool-versions/.nvmrc are still honored under mise before trusting the switch.
  • Back up shell config (~/.zshrc, ~/.config/fish/config.fish) before neutralizing a manager.
  • Do not remove nvm/asdf/pyenv/node@x until mise is verified in a clean env in BOTH shells — a half-migrated PATH silently falls back to the old manager (or to no runtime at all).
  • Confirm before editing shell config, mise use -g (changes global default), or any uninstall.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.