Skip to content

Jarvy vs asdf

A comparison of Jarvy and asdf for developer environment management.

Quick Comparison

Feature Jarvy asdf
Purpose Full environment provisioning Runtime version management
Scope 40+ tools (runtimes, apps, CLIs) Language runtimes via plugins
Config Format jarvy.toml .tool-versions
Mechanism Native package managers Shims + 500+ plugins
Version Switching Single version per tool Multiple versions, per-directory
Language Rust Shell scripts

When to Choose Jarvy

  • New machine setup - Bootstrap a fresh laptop with everything needed
  • Full application installs - Docker, VS Code, iTerm2, Terraform
  • Infrastructure tools - AWS CLI, Packer, OpenTofu, k6
  • Team onboarding - Single command provisions entire environment
  • Cross-platform - Native support for macOS, Linux, Windows
  • No shim overhead - Tools installed directly, no command interception

When to Choose asdf

  • Version switching - Different Node versions per project
  • Per-directory activation - Automatic switching on cd
  • Plugin ecosystem - 500+ community plugins for any tool
  • Existing adoption - Team already uses .tool-versions files
  • Shell integration - Deep integration with bash/zsh
  • Build from source - Often compiles tools rather than downloading binaries

Key Differentiators

Jarvy's Approach

  • Uses native package managers (Homebrew, apt, winget)
  • Installs full applications, not just runtimes
  • Global installation - tools available everywhere
  • Simple TOML configuration
  • No shims, no shell hooks

asdf's Approach

  • Plugin architecture for any tool
  • Shims intercept commands and route to correct version
  • Per-directory version files
  • Community-maintained plugins
  • Shell-based implementation

Can They Work Together?

Yes. They solve different problems:

  1. Jarvy for machine provisioning - One-time setup of Docker, VS Code, Terraform, asdf itself
  2. asdf for runtime management - Per-project Node, Python, Ruby versions
# jarvy.toml
[tools]
docker = "latest"
vscode = "latest"
terraform = "1.5"
asdf = "latest"  # Jarvy installs asdf

# .tool-versions (asdf manages these)
nodejs 20.10.0
python 3.12.0
ruby 3.3.0

Migration Considerations

From asdf to Jarvy: - Keep asdf for runtime version management - Use Jarvy for everything else (Docker, VS Code, CLI tools) - Jarvy doesn't replace asdf's version switching

From Jarvy to asdf: - asdf won't install Docker, VS Code, or GUI applications - asdf focuses on runtimes, not full environment provisioning - Consider using both tools together