Skip to content
English
Dashboard

CLI Installation

The SnapVM CLI (snapvm) lets you work with machines from your local terminal. Use it to sign in, list or create machines, run commands, open an interactive shell, and manage checkpoints without leaving your command line.

Install the latest CLI on macOS or Linux:

curl -fsSL https://snapvm.dev/install.sh | sh

The install script:

  • Detects your operating system and CPU architecture
  • Downloads the matching SnapVM CLI release
  • Verifies the release checksum
  • Installs the binary as snapvm

By default, the binary is installed to /usr/local/bin/snapvm. The script may ask for sudo if your user cannot write to that directory.

If you prefer a user-writable directory, set SNAPVM_INSTALL_DIR before running the script:

export SNAPVM_INSTALL_DIR="$HOME/.local/bin"
curl -fsSL https://snapvm.dev/install.sh | sh

Make sure the directory is on your PATH:

export PATH="$HOME/.local/bin:$PATH"

Add that PATH update to your shell profile if you want it to persist across terminal sessions.

Confirm that your shell can find the CLI:

which snapvm
snapvm --version
snapvm --help

You should see a snapvm path, a version, and the command list.

Update an existing CLI installation:

snapvm update

Use snapvm use to make a machine the default for commands run from the current project directory:

snapvm use frontend-lab

The CLI stores this local context in a .snapvm file. Add .snapvm to your .gitignore if machine selection should remain personal to each developer.

Remove the local context when you no longer want the directory to target that machine:

snapvm use --unset

Check where the CLI was installed:

which snapvm

If the command is not found, add the install directory to your PATH. For the default install location:

export PATH="/usr/local/bin:$PATH"

The default install location may require administrator permission. Either allow the script to use sudo, or install into a user-writable directory with SNAPVM_INSTALL_DIR.

The install script supports macOS and Linux on amd64/x86_64 and arm64/aarch64. If your platform is not supported, use a supported machine or contact your workspace administrator.