Copyright 2016-2025 Caleb Evans
Released under the MIT license

Alfred Workflow Packager is a command-line utility which makes the process of packaging and exporting an Alfred workflow incredibly quick and easy. The utility supports Alfred 3 and up, on projects running Python 3 (Python 2 is no longer supported).

You can install the utility via pip3, either globally or within a virtualenv:

pip3 install alfred-workflow-packager

Once you've installed AWP, you must configure it for every project where you wish to use it. To do so, create a packager.json file in the root directory of your project; this file configures AWP for that particular project.

{
  "export_files": [
    "Fruit.alfredworkflow"
  ],
  "bundle_id": "com.yourname.fruit",
  "readme": "README.txt",
  "resources": [
      "icon.png",
      "src/*.py",
      "src/data/*.json"
  ]
}

The paths of the exported workflows (relative to your project directory).

The unique bundle ID of your workflow. You must have one set in the installed workflow, or AWP will not be able to find your workflow when packaging.

A list of zero or more files/folder patterns representing files or folders to copy from your project to the installed workflow. The directory structures and filenames are preserved when copying.

Local project:

- icon.png
- fruit
    - apple.py
    - banana.applescript
    - orange.php

Installed workflow (before running utility):

- info.plist
- special_file.json

packager.json resources:

[
    "icon.png",
    "fruit/*.json"
]

Installed workflow (after running utility):

- info.plist
- icon.png
- special_file.json
- fruit
    - apple.py
    - banana.applescript
    - orange.php

Note that files and folders already present in the installed workflow are not touched if they are not in the resources list.

The path to the README file to use for this workflow; the About this Workflow field in your workflow is populated with the contents of this file.

You can run the utility via the awp command:

awp

Running awp will copy those project resources listed in packager.json to the installed workflow (in their respective locations), but only if their contents or permissions have changed. If you ever need to ignore this equality check, you can force the copying of all files/directories by passing --force / -f.

awp --force

awp -f

Passing the --version option (also -v) to awp allows you to set the version of the installed workflow directly. I highly recommend using semantic versioning to version your workflow releases.

awp --version 1.2.0

awp -v 1.2.0

When you're pleased with your work and you're ready to publish a new release, you can export the installed workflow to your project directory by passing the --export flag (or -e) to awp.

awp --export

awp -e

Note that you can set the version and export the workflow simultaneously:

awp -v 1.2.0 -e

New in AWP v1.1.0: If you wish to temporarily export the workflow to a different file (different from export_files in packager.json), you can pass one or more optional paths to --export:

awp -v 1.3.0-beta.1 -e ~/Desktop/fruit-beta-alfred-5.alfredworkflow

The last important step is to update any script objects in your workflow (i.e. objects of type Script Filter, Run Script, etc.) to reference the files copied to the installed workflow directory.

You should set the Language to /bin/bash and use the appropriate shell command to call your script. Use "$@" if your input is passed as argv, or "{query}" if your input is passed as {query}.

/usr/bin/python3 -m fruit.apple "$@"

/usr/bin/python3 -m fruit.apple "{query}"

/usr/bin/osascript fruit/banana.applescript "$@"

/usr/bin/osascript fruit/banana.applescript "{query}"

/usr/bin/php fruit/orange.php "$@"

/usr/bin/php fruit/orange.php "{query}"