Skip to content

Stack's script interpreter

Stack offers a very useful feature for running files: a script interpreter. For too long have Haskellers felt shackled to bash or Python because it's just too hard to create reusable source-only Haskell scripts. Stack attempts to solve that.

You can use stack <file_name> to execute a Haskell source file. Usually, the Stack command to be applied is specified using a special Haskell comment (the Stack interpreter options comment) at the start of the source file. That command is most often stack script but it can be, for example, stack runghc. If there is no Stack interpreter options comment, Stack will warn that one was expected.

An example will be easiest to understand. Consider the Haskell source file turtle-example.hs with contents:

#!/usr/bin/env stack
-- stack script --snapshot lts-22.21 --package turtle
{-# LANGUAGE OverloadedStrings #-}
import Turtle (echo)
main = echo "Hello World!"

The first line beginning with the 'shebang' (#!) tells Unix to use Stack as a script interpreter, if the file's permissions mark it as executable. A shebang line is limited to a single argument, here stack.

The file's permissions can be set with command chmod and then it can be run:

chmod +x turtle-example.hs


On macOS:

  • Avoid {-# LANGUAGE CPP #-} in Stack scripts; it breaks the shebang line (GHC #6132)

  • Use a compiled executable, not another script, in the shebang line. Eg #!/usr/bin/env runhaskell will work but #!/usr/local/bin/runhaskell would not.

Alternatively, the script can be run with command:

stack turtle-example.hs

The first line beginning with the 'shebang' (#!) has a meaning on Unix-like operating systems but will be ignored by PowerShell. It can be omitted on Windows. The script can be run with command:

stack turtle-example.hs

In both cases, the command yields:

Hello World!

the first time after a little delay (as GHC is downloaded, if necessary, and dependencies are built) and subsequent times more promptly (as the runs are able to reuse everything already built).

The second line of the source code is the Stack interpreter options comment. In this example, it specifies the stack script command with the options of a LTS Haskell 22.21 snapshot (--snapshot lts-22.21) and ensuring the turtle package is available (--package turtle). The version of the package will be that in the specified snapshot (lts-22.21 provides turtle-1.6.2).

Arguments and interpreter options and arguments

Arguments for the script can be specified on the command line after the file name: stack <file_name> <arg1> <arg2> ....

The Stack interpreter options comment must specify what would be a single valid Stack command at the command line if the file name were included as an argument, starting with stack. It can include -- followed by arguments. In particular, the Stack command stack <arg1> MyScript.hs <arg4> with Stack interpreter options comment:

-- stack <arg2> <command> <arg3> -- <arg5>

is equivalent to the following command at the command line:

stack <arg1> <arg2> <command> <arg3> -- MyScript.hs <arg4> <arg5>

The Stack interpreter options comment must be the first line of the file, unless a shebang line is the first line, when the comment must be the second line. The comment must start in the first column of the line.

When many options are needed, a block style comment that splits the command over more than one line may be more convenient and easier to read.

For example, the command stack MyScript.hs arg1 arg2 with MyScript.hs:

#!/usr/bin/env stack
{- stack script
   --snapshot lts-22.21
   +RTS -s -RTS
import Data.List (intercalate)
import System.Environment (getArgs)
import Turtle (echo, fromString)

main = do
  args <- getArgs
  echo $ fromString $ intercalate ", " args

is equivalent to the following command at the command line:

stack script --snapshot lts-22.21 -- MyScript.hs arg1 arg2 +RTS -s -RTS

where +RTS -s -RTS are some of GHC's runtime system (RTS) options.

Just-in-time compilation

As with using stack script at the command line, you can pass the --compile flag to make Stack compile the script, and then run the compiled executable. Compilation is done quickly, without optimization. To compile with optimization, pass the --optimize flag instead. Compilation is done only if needed; if the executable already exists, and is newer than the script, Stack just runs the executable directly.

This feature can be good for speed (your script runs faster) and also for durability (the executable remains runnable even if the script is disturbed, eg due to changes in your installed GHC/snapshots, changes to source files during git bisect, etc.)

Using multiple packages

As with using stack script at the command line, you can also specify multiple packages, either with multiple --package options, or by providing a comma or space separated list. For example:

#!/usr/bin/env stack
{- stack script
   --snapshot lts-22.21
   --package turtle
   --package "stm async"
   --package http-client,http-conduit

Stack configuration for scripts

When using the stack script command, as when using it at the command line, any project-level configuration file (stack.yaml, by default) (including in the global-project directory in the Stack root), including any specified by the options to the stack script command itself, is ignored.


Non-project level configuration options in global configuration files (config.yaml), are not ignored by the stack script command. Such options may be useful if allow-newer and/or allow-newer-deps are required.

When using the stack runghc command, as when using it at the command line, if the current working directory is inside a project, then that project's project-level configuration file is effective when running the script. Otherwise the script uses the project-level configuration file in the global-project directory in the Stack root.

Testing scripts

You can use the flag --script-no-run-compile on the command line to enable (it is disabled by default) the use of the --no-run option with stack script (and forcing the --compile option). The flag may help test that scripts compile in CI (continuous integration).

For example, consider the following simple script, in a file named Script.hs, which makes use of the joke package acme-missiles:

{- stack script
   --snapshot lts-22.21
   --package acme-missiles
import Acme.Missiles (launchMissiles)

main :: IO ()
main = launchMissiles

The command stack --script-no-run-compile Script.hs then behaves as if the command stack script --snapshot lts-22.21 --package acme-missiles --no-run --compile -- Script.hs had been given. Script.hs is compiled (without optimisation) and the resulting executable is not run: no missiles are launched in the process!

Writing independent and reliable scripts

The stack script command will automatically:

  • Install GHC and libraries, if missing. stack script behaves as if the --install-ghc flag had been passed at the command line.
  • Require that all packages used be explicitly stated on the command line.

This ensures that your scripts are independent of any prior deployment specific configuration, and are reliable by using exactly the same version of all packages every time it runs so that the script does not break by accidentally using incompatible package versions.

In earlier versions of Stack, the stack runghc command was used for scripts and can still be used in that way. In order to achieve the same effect with the stack runghc command, you can do the following:

  1. Use the --install-ghc option to install the compiler automatically
  2. Explicitly specify all packages required by the script using the --package option. Use -hide-all-packages GHC option to force explicit specification of all packages.
  3. Use the --snapshot Stack option to ensure a specific GHC version and package set is used.

It is possible for a project-level configuration file to affect stack runghc. For that reason, stack script is strongly recommended. For those curious, here is an example with stack runghc:

#!/usr/bin/env stack
{- stack
  --snapshot lts-22.21
  --package base
  --package turtle

The stack runghc command is still useful, especially when you're working on a project and want to access the package databases and configurations used by that project. See the next section for more information on configuration files.

Loading scripts in GHCi

Sometimes you want to load your script in GHCi to play around with your program. In those cases, you can use exec ghci option in the script to achieve it. Here is an example:

#!/usr/bin/env stack
{- stack
   exec ghci
   --snapshot lts-22.21
   --package turtle