raiku run executes a shell command inside the cached directory of an installed package, similar to npx. Useful for running package-specific tools or scripts.
Use this file to discover all available pages before exploring further.
raiku run lets you execute any shell command from within the cached directory of an installed package — similar in spirit to npx in the Node.js ecosystem. Rather than navigating to the cache directory manually, you pass the package name and the command you want to run; Raiku resolves the cache path, sets up a clean environment with package metadata variables, and executes the command with the package directory as the working directory. This is useful for invoking package-specific scripts, running one-off builds, inspecting package internals, or testing commands without permanently modifying your shell environment.
The command is resolved against the package’s cache directory, not your current working directory.
If multiple versions of the same package are installed, the latest version is used unless --version is specified.
The exit code of raiku run matches the exit code of the subprocess exactly, making it safe to use in scripts.
On Windows, the command is run without a shell wrapper (shell=False). On Unix, it is also run directly without shell expansion.
Verbose mode (raiku --verbose run ...) prints the resolved working directory and full command before execution.
The package must be installed before you can run commands inside it. Use raiku install PACKAGE first. If the package is listed in the index but not cached locally, raiku run will exit with an error.
Use raiku run PACKAGE ls src/ to quickly inspect the contents of a package’s src/ directory without navigating to the cache path manually. This is handy when exploring a newly installed package.