local development server that aims to make using browserify fast and fun

npm install beefy
86 downloads in the last day
306 downloads in the last week
1 359 downloads in the last month


a local development server designed to work with browserify.


  • can live reload your browser when your code changes (if you want)
  • works with whatever version of browserify; globally installed or locally installed to node_modules/browserify.
  • will spit compile errors out into the browser so you don't have that 1-2 seconds of cognitive dissonance and profound ennui that follows refreshing the page only to get a blank screen.
  • will spit out a default index.html for missing routes so you don't need to even muck about with HTML to get started
  • serves up static files with grace and aplomb (and also appropriate mimetypes)
  • makes it easy to sanity check your testling ci tape test suite.
  • loves you, unconditionally

how do I get it?

npm install -g beefy; and if you want to always have a browserify available for beefy to use, npm install -g browserify.


$ cd directory/you/want/served
$ beefy path/to/thing/you/want/browserified.js PORT -- [browserify args]


the path to the file you want browserified. can be just a normal node module. you can also alias it: path/to/file.js:bundle.js if you want -- so all requests to bundle.js will browserify path/to/file.js. this is helpful for when you're writing gh-pages-style sites that already have an index.html, and expect the bundle to be pregenerated and available at a certain path.

--browserify command

--bundler command

use command instead of browserify or ./node_modules/.bin/browserify.

in theory, you could even get this working with r.js, but that would probably be scary and bats would fly out of it. but it's there if you need it! if you want to use r.js with beefy, you'll need a config that can write the resulting bundle to stdout, and you can run beefy with beefy :output-url.js --bundler r.js -- -o config.js.


enable live reloading. this'll start up a sideband server and an fs watch on the current working directory -- if you save a file, your browser will refresh.

if you're not using the generated index file, put the following script tag above all of your other JS:

    <script src="/-/live-reload.js"></script>

--cwd dir

serve files as if running from dir.


turn off browserify source map output. by default, beefy automatically inserts -d into the browserify args -- this turns that behavior off.


automatically discover a port and open it using your default browser.


beefy exports one function which returns a http server created from http.createServer()

beefy(cwd, browserify_path, browserify_args, entry_points, live_reload, log, custom_handler)

  • cwd (string) root folder beefy uses for serving content. this folder is also watched if the live_reload parameter is set.
  • browserify_path (string) command to execute when browserifying the code. use 'browserify' for standard behavior.
  • browserify_args (array of strings) arguments to the browserify command. use e.g. [ '-d' ] for debug mode.
  • entry_points (object) dictionary for your entry points and corresponding file to browserify. see example below.
  • live_reload (boolean) enable live reload if set
  • log (function) optional logging callback. see signature below.
  • custom_handler (function) optional custom request handler. return falsy in handler to delegate back to beefy.
var beefy = require('beefy')
var entry_points = { 'bundle.js': 'path/to/some/js/file.js' }
var server = beefy('path/to/wwwroot', 'browserify', [ '-d' ], entry_points, true, log,

function log(code, time, bytesize, logged_pathname, color) {}

function custom_handler(req, resp) {
  if (req.url == '/foo') {
    // custom handling of '/foo'
    return true
  // delegate back to beefy

the server object is patched with a reload() method which allows you to reload clients programmatically:

var watchr = require('watchr'){
    path: '/some/other/path/not/watched/by/beefy'
  , listener: function (event, file, stat_now, stat_then) {
      // do stuff ..

the fake index

by default, if you get a URL that doesn't exist (with an Accept header that has html in it someplace), you'll get the "fake index." this page is setup so that it automatically includes both the live reload script (if it's enabled) and the path you want browserified.



npm loves you