diff --git a/scrapy/cmdline.py b/scrapy/cmdline.py
index c7e859515..8d1a31d16 100644
--- a/scrapy/cmdline.py
+++ b/scrapy/cmdline.py
@@ -6,11 +6,10 @@ import optparse
 import cProfile
 
 import scrapy
-from scrapy import log
 from scrapy.crawler import CrawlerProcess
 from scrapy.xlib import lsprofcalltree
 from scrapy.conf import settings
-from scrapy.command import ScrapyCommand
+from scrapy.exceptions import UsageError
 
 def _find_commands(dir):
     try:
@@ -19,47 +18,61 @@ def _find_commands(dir):
     except OSError:
         return []
 
-def _get_commands_from_module(module):
+def _get_commands_from_module(module, inproject):
     d = {}
     mod = __import__(module, {}, {}, [''])
     for cmdname in _find_commands(mod.__path__[0]):
         modname = '%s.%s' % (module, cmdname)
         command = getattr(__import__(modname, {}, {}, [cmdname]), 'Command', None)
         if callable(command):
-            d[cmdname] = command()
+            if inproject or not command.requires_project:
+                d[cmdname] = command()
         else:
-            print 'WARNING: Module %r does not define a Command class' % modname
+            raise RuntimeError("Module %r does not define a Command class" % modname)
     return d
 
-def _get_commands_dict():
-    cmds = _get_commands_from_module('scrapy.commands')
+def _get_commands_dict(inproject):
+    cmds = _get_commands_from_module('scrapy.commands', inproject)
     cmds_module = settings['COMMANDS_MODULE']
     if cmds_module:
-        cmds.update(_get_commands_from_module(cmds_module))
+        cmds.update(_get_commands_from_module(cmds_module, inproject))
     return cmds
 
-def _get_command_name(argv):
+def _pop_command_name(argv):
+    i = 0
     for arg in argv[1:]:
         if not arg.startswith('-'):
+            del argv[i]
             return arg
+        i += 1
 
-def _print_usage(inside_project):
-    if inside_project:
+def _print_header(inproject):
+    if inproject:
         print "Scrapy %s - project: %s\n" % (scrapy.__version__, \
             settings['BOT_NAME'])
     else:
         print "Scrapy %s - no active project\n" % scrapy.__version__
+
+def _print_commands(inproject):
+    _print_header(inproject)
     print "Usage:"
     print "  scrapy <command> [options] [args]\n"
     print "Available commands:"
-    cmds = _get_commands_dict()
+    cmds = _get_commands_dict(inproject)
     for cmdname, cmdclass in sorted(cmds.iteritems()):
-        if inside_project or not cmdclass.requires_project:
-            print "  %-13s %s" % (cmdname, cmdclass.short_desc())
+        print "  %-13s %s" % (cmdname, cmdclass.short_desc())
     print
-    print 'Use "scrapy <command> -h" for more info about a command'
+    print 'Use "scrapy <command> -h" to see more info about a command'
 
-def check_deprecated_scrapy_ctl(argv):
+def _print_unknown_command(cmdname, inproject):
+    _print_header(inproject)
+    print "Unknown command: %s\n" % cmdname
+    print 'Use "scrapy" to see available commands' 
+    if not inproject:
+        print
+        print "More commands are available in project mode"
+
+def _check_deprecated_scrapy_ctl(argv, inproject):
     """Check if Scrapy was called using the deprecated scrapy-ctl command and
     warn in that case, also creating a scrapy.cfg if it doesn't exist.
     """
@@ -68,7 +81,7 @@ def check_deprecated_scrapy_ctl(argv):
     import warnings
     warnings.warn("`scrapy-ctl.py` command-line tool is deprecated and will be removed in Scrapy 0.11, use `scrapy` instead",
         DeprecationWarning, stacklevel=3)
-    if settings.settings_module:
+    if inproject:
         projpath = os.path.abspath(os.path.dirname(os.path.dirname(settings.settings_module.__file__)))
         cfg_path = os.path.join(projpath, 'scrapy.cfg')
         if not os.path.exists(cfg_path):
@@ -77,72 +90,64 @@ def check_deprecated_scrapy_ctl(argv):
                 f.write("[default]" + os.linesep)
                 f.write("settings = %s" % settings.settings_module_path + os.linesep)
 
+def _run_print_help(parser, func, *a, **kw):
+    try:
+        func(*a, **kw)
+    except UsageError, e:
+        if str(e):
+            parser.error(str(e))
+        if e.print_help:
+            parser.print_help()
+        sys.exit(2)
+
 def execute(argv=None):
     if argv is None:
         argv = sys.argv
     crawler = CrawlerProcess(settings)
     crawler.install()
-    check_deprecated_scrapy_ctl(argv) # TODO: remove for Scrapy 0.11
-    cmds = _get_commands_dict()
-    cmdname = _get_command_name(argv)
+    inproject = bool(settings.settings_module)
+    _check_deprecated_scrapy_ctl(argv, inproject) # TODO: remove for Scrapy 0.11
+    cmds = _get_commands_dict(inproject)
+    cmdname = _pop_command_name(argv)
     parser = optparse.OptionParser(formatter=optparse.TitledHelpFormatter(), \
-        conflict_handler='resolve', add_help_option=False)
-    if cmdname in cmds:
-        cmd = cmds[cmdname]
-        cmd.add_options(parser)
-        opts, args = parser.parse_args(args=argv[1:])
-        cmd.process_options(args, opts)
-        parser.usage = "%%prog %s %s" % (cmdname, cmd.syntax())
-        parser.description = cmd.long_desc()
-        if cmd.requires_project and not settings.settings_module:
-            print "Error running: scrapy %s\n" % cmdname
-            print "Cannot find project settings module in python path: %s" % \
-                settings.settings_module_path
-            sys.exit(1)
-        if opts.help:
-            parser.print_help()
-            sys.exit()
-    elif not cmdname:
-        cmd = ScrapyCommand()
-        cmd.add_options(parser)
-        opts, args = parser.parse_args(args=argv)
-        cmd.process_options(args, opts)
-        _print_usage(settings.settings_module)
-        sys.exit(2)
-    else:
-        print "Unknown command: %s\n" % cmdname
-        print 'Use "scrapy -h" for help' 
+        conflict_handler='resolve')
+    if not cmdname:
+        _print_commands(inproject)
+        sys.exit(0)
+    elif cmdname not in cmds:
+        _print_unknown_command(cmdname, inproject)
         sys.exit(2)
 
+    cmd = cmds[cmdname]
+    parser.usage = "scrapy %s %s" % (cmdname, cmd.syntax())
+    parser.description = cmd.long_desc()
     settings.defaults.update(cmd.default_settings)
-    del args[0]  # remove command name from args
-    log.start()
     cmd.set_crawler(crawler)
-    ret = _run_command(cmd, args, opts)
-    if ret is False:
-        parser.print_help()
+    cmd.add_options(parser)
+    opts, args = parser.parse_args(args=argv[1:])
+    _run_print_help(parser, cmd.process_options, args, opts)
+    _run_print_help(parser, _run_command, cmd, args, opts)
 
 def _run_command(cmd, args, opts):
     if opts.profile or opts.lsprof:
-        return _run_command_profiled(cmd, args, opts)
+        _run_command_profiled(cmd, args, opts)
     else:
-        return cmd.run(args, opts)
+        cmd.run(args, opts)
 
 def _run_command_profiled(cmd, args, opts):
     if opts.profile:
-        log.msg("writing cProfile stats to %r" % opts.profile)
+        sys.stderr.write("scrapy: writing cProfile stats to %r\n" % opts.profile)
     if opts.lsprof:
-        log.msg("writing lsprof stats to %r" % opts.lsprof)
+        sys.stderr.write("scrapy: writing lsprof stats to %r\n" % opts.lsprof)
     loc = locals()
     p = cProfile.Profile()
-    p.runctx('ret = cmd.run(args, opts)', globals(), loc)
+    p.runctx('cmd.run(args, opts)', globals(), loc)
     if opts.profile:
         p.dump_stats(opts.profile)
     k = lsprofcalltree.KCacheGrind(p)
     if opts.lsprof:
         with open(opts.lsprof, 'w') as f:
             k.output(f)
-    return loc['ret']
 
 if __name__ == '__main__':
     execute()
diff --git a/scrapy/command.py b/scrapy/command.py
index 39bdb325d..a9c447e38 100644
--- a/scrapy/command.py
+++ b/scrapy/command.py
@@ -9,8 +9,10 @@ import sys
 from optparse import OptionGroup
 
 import scrapy
+from scrapy import log
 from scrapy.conf import settings
 from scrapy.utils.conf import arglist_to_dict
+from scrapy.exceptions import UsageError
 
 class ScrapyCommand(object):
 
@@ -24,6 +26,8 @@ class ScrapyCommand(object):
 
     @property
     def crawler(self):
+        if not log.started:
+            log.start()
         self._crawler.configure()
         return self._crawler
 
@@ -58,10 +62,6 @@ class ScrapyCommand(object):
         Populate option parse with options available for this command
         """
         group = OptionGroup(parser, "Global Options")
-        group.add_option("-h", "--help", action="store_true", dest="help", \
-            help="print command help and options")
-        group.add_option("--version", action="store_true", dest="version", \
-            help="print Scrapy version and exit")
         group.add_option("--logfile", dest="logfile", metavar="FILE", \
             help="log file. if omitted stderr will be used")
         group.add_option("-L", "--loglevel", dest="loglevel", metavar="LEVEL", \
@@ -77,23 +77,13 @@ class ScrapyCommand(object):
             help="write process ID to FILE")
         group.add_option("--set", dest="set", action="append", default=[], metavar="NAME=VALUE", \
             help="set/override setting (may be repeated)")
-        group.add_option("--settings", dest="settings", metavar="MODULE",
-            help="python path to the Scrapy project settings")
         parser.add_option_group(group)
         
     def process_options(self, args, opts):
-        if opts.settings:
-            settings.set_settings_module(opts.settings)
-
         try:
             settings.overrides.update(arglist_to_dict(opts.set))
         except ValueError:
-            sys.stderr.write("Invalid --set value, use --set NAME=VALUE\n")
-            sys.exit(2)
-
-        if opts.version:
-            print "Scrapy %s" % scrapy.__version__
-            sys.exit()
+            raise UsageError("Invalid --set value, use --set NAME=VALUE", print_help=False)
 
         if opts.logfile:
             settings.overrides['LOG_ENABLED'] = True
@@ -115,4 +105,3 @@ class ScrapyCommand(object):
         Entry point for running commands
         """
         raise NotImplementedError
-
diff --git a/scrapy/commands/crawl.py b/scrapy/commands/crawl.py
index bc99c8f23..d27ba7f9b 100644
--- a/scrapy/commands/crawl.py
+++ b/scrapy/commands/crawl.py
@@ -6,6 +6,7 @@ from scrapy.conf import settings
 from scrapy.http import Request
 from scrapy.utils.url import is_url
 from scrapy.utils.conf import arglist_to_dict
+from scrapy.exceptions import UsageError
 
 from collections import defaultdict
 
@@ -33,8 +34,7 @@ class Command(ScrapyCommand):
         try:
             opts.spargs = arglist_to_dict(opts.spargs)
         except ValueError:
-            sys.stderr.write("Invalid -a value, use -a NAME=VALUE\n")
-            sys.exit(2)
+            raise UsageError("Invalid -a value, use -a NAME=VALUE", print_help=False)
         if opts.nofollow:
             settings.overrides['CRAWLSPIDER_FOLLOW_LINKS'] = False
 
diff --git a/scrapy/commands/fetch.py b/scrapy/commands/fetch.py
index 980348b12..4d1abe14e 100644
--- a/scrapy/commands/fetch.py
+++ b/scrapy/commands/fetch.py
@@ -5,6 +5,7 @@ from scrapy.command import ScrapyCommand
 from scrapy.http import Request
 from scrapy.spider import BaseSpider
 from scrapy.utils.url import is_url
+from scrapy.exceptions import UsageError
 
 class Command(ScrapyCommand):
 
@@ -35,7 +36,7 @@ class Command(ScrapyCommand):
 
     def run(self, args, opts):
         if len(args) != 1 or not is_url(args[0]):
-            return False
+            raise UsageError()
         cb = lambda x: self._print_response(x, opts)
         request = Request(args[0], callback=cb, dont_filter=True)
 
diff --git a/scrapy/commands/genspider.py b/scrapy/commands/genspider.py
index 35bb0f23e..e3f869f4c 100644
--- a/scrapy/commands/genspider.py
+++ b/scrapy/commands/genspider.py
@@ -7,6 +7,7 @@ import scrapy
 from scrapy.command import ScrapyCommand
 from scrapy.conf import settings
 from scrapy.utils.template import render_templatefile, string_camelcase
+from scrapy.exceptions import UsageError
 
 def sanitize_module_name(module_name):
     """Sanitize the given module name, by replacing dashes and points
@@ -54,7 +55,7 @@ class Command(ScrapyCommand):
                 print open(template_file, 'r').read()
             return
         if len(args) != 2:
-            return False
+            raise UsageError()
 
         name, domain = args[0:2]
         module = sanitize_module_name(name)
diff --git a/scrapy/commands/parse.py b/scrapy/commands/parse.py
index 839b226ab..68f727e48 100644
--- a/scrapy/commands/parse.py
+++ b/scrapy/commands/parse.py
@@ -4,6 +4,7 @@ from scrapy.item import BaseItem
 from scrapy.utils import display
 from scrapy.utils.spider import iterate_spider_output
 from scrapy.utils.url import is_url
+from scrapy.exceptions import UsageError
 from scrapy import log
 
 class Command(ScrapyCommand):
@@ -93,7 +94,7 @@ class Command(ScrapyCommand):
 
     def run(self, args, opts):
         if not len(args) == 1 or not is_url(args[0]):
-            return False
+            raise UsageError()
         response, spider = self.get_response_and_spider(args[0], opts)
         if not response:
             return
diff --git a/scrapy/commands/queue.py b/scrapy/commands/queue.py
index 03379d8f6..fb6ee594e 100644
--- a/scrapy/commands/queue.py
+++ b/scrapy/commands/queue.py
@@ -1,6 +1,6 @@
 from scrapy.command import ScrapyCommand
 from scrapy.commands import runserver
-from scrapy.utils.misc import load_object
+from scrapy.exceptions import UsageError
 from scrapy.conf import settings
 
 class Command(runserver.Command):
@@ -23,7 +23,7 @@ class Command(runserver.Command):
 
     def run(self, args, opts):
         if len(args) < 1:
-            return False
+            raise UsageError()
         cmd = args[0]
 
         botname = settings['BOT_NAME']
@@ -31,7 +31,7 @@ class Command(runserver.Command):
 
         if cmd == 'add':
             if len(args) < 2:
-                return False
+                raise UsageError()
             msg = dict(x for x in [x.split('=', 1) for x in opts.spargs])
             for x in args[1:]:
                 msg.update(name=x)
@@ -44,4 +44,4 @@ class Command(runserver.Command):
             queue.clear()
             print "Cleared %s queue" % botname
         else:
-            return False
+            raise UsageError()
diff --git a/scrapy/commands/runspider.py b/scrapy/commands/runspider.py
index 84253c66d..e9804e061 100644
--- a/scrapy/commands/runspider.py
+++ b/scrapy/commands/runspider.py
@@ -1,9 +1,9 @@
 import sys
 import os
 
-from scrapy import log
 from scrapy.utils.spider import iter_spider_classes
 from scrapy.command import ScrapyCommand
+from scrapy.exceptions import UsageError
 
 def _import_file(filepath):
     abspath = os.path.abspath(filepath)
@@ -33,22 +33,22 @@ class Command(ScrapyCommand):
     def long_desc(self):
         return "Run the spider defined in the given file"
 
+    def process_options(self, args, opts):
+        ScrapyCommand.process_options(self, args, opts)
+
     def run(self, args, opts):
         if len(args) != 1:
-            return False
+            raise UsageError()
         filename = args[0]
         if not os.path.exists(filename):
-            log.msg("File not found: %s\n" % filename, log.ERROR)
-            return
+            raise UsageError("File not found: %s\n" % filename)
         try:
             module = _import_file(filename)
         except (ImportError, ValueError), e:
-            log.msg("Unable to load %r: %s\n" % (filename, e), log.ERROR)
-            return
+            raise UsageError("Unable to load %r: %s\n" % (filename, e))
         spclasses = list(iter_spider_classes(module))
         if not spclasses:
-            log.msg("No spider found in file: %s\n" % filename, log.ERROR)
-            return
+            raise UsageError("No spider found in file: %s\n" % filename)
         spider = spclasses.pop()()
         # schedule spider and start engine
         self.crawler.queue.append_spider(spider)
diff --git a/scrapy/commands/startproject.py b/scrapy/commands/startproject.py
index b398f78e0..0b39630ac 100644
--- a/scrapy/commands/startproject.py
+++ b/scrapy/commands/startproject.py
@@ -8,6 +8,7 @@ import scrapy
 from scrapy.command import ScrapyCommand
 from scrapy.utils.template import render_templatefile, string_camelcase
 from scrapy.utils.py26 import ignore_patterns, copytree
+from scrapy.exceptions import UsageError
 
 TEMPLATES_PATH = join(scrapy.__path__[0], 'templates', 'project')
 
@@ -33,7 +34,7 @@ class Command(ScrapyCommand):
 
     def run(self, args, opts):
         if len(args) != 1:
-            return False
+            raise UsageError()
         project_name = args[0]
         if not re.search(r'^[_a-zA-Z]\w*$', project_name):
             print 'Error: Project names must begin with a letter and contain only\n' \
diff --git a/scrapy/commands/version.py b/scrapy/commands/version.py
new file mode 100644
index 000000000..4d4557357
--- /dev/null
+++ b/scrapy/commands/version.py
@@ -0,0 +1,10 @@
+import scrapy
+from scrapy.command import ScrapyCommand
+
+class Command(ScrapyCommand):
+
+    def short_desc(self):
+        return "Print Scrapy version"
+
+    def run(self, args, opts):
+        print "Scrapy %s" % scrapy.__version__
diff --git a/scrapy/exceptions.py b/scrapy/exceptions.py
index 48204f148..3294dc4da 100644
--- a/scrapy/exceptions.py
+++ b/scrapy/exceptions.py
@@ -38,3 +38,11 @@ class DropItem(Exception):
 class NotSupported(Exception):
     """Indicates a feature or method is not supported"""
     pass
+
+# Commands
+
+class UsageError(Exception):
+    """To indicate a command-line usage error"""
+    def __init__(self, *a, **kw):
+        self.print_help = kw.pop('print_help', True)
+        super(UsageError, self).__init__(*a, **kw)
diff --git a/scrapy/tests/test_cmdline/__init__.py b/scrapy/tests/test_cmdline/__init__.py
index 697ebb512..3588436b9 100644
--- a/scrapy/tests/test_cmdline/__init__.py
+++ b/scrapy/tests/test_cmdline/__init__.py
@@ -23,11 +23,6 @@ class CmdlineTest(unittest.TestCase):
         self.assertEqual(self._execute('settings', '--get', 'TEST1'), \
             'default + loaded + started')
 
-    def test_override_settings_using_settings_arg(self):
-        self.assertEqual(self._execute('settings', '--get', 'TEST1', \
-            '--settings', 'scrapy.tests.test_cmdline.settings2'), \
-            'override + loaded + started')
-
     def test_override_settings_using_set_arg(self):
         self.assertEqual(self._execute('settings', '--get', 'TEST1', '--set', 'TEST1=override'), \
             'override + loaded + started')
diff --git a/scrapy/tests/test_cmdline/settings2.py b/scrapy/tests/test_cmdline/settings2.py
deleted file mode 100644
index 4b5ba9e16..000000000
--- a/scrapy/tests/test_cmdline/settings2.py
+++ /dev/null
@@ -1,5 +0,0 @@
-EXTENSIONS = [
-    'scrapy.tests.test_cmdline.extensions.TestExtension'
-]
-
-TEST1 = 'override'
diff --git a/scrapy/tests/test_commands.py b/scrapy/tests/test_commands.py
index 7ece8a62b..47f67338a 100644
--- a/scrapy/tests/test_commands.py
+++ b/scrapy/tests/test_commands.py
@@ -69,7 +69,7 @@ class GenspiderCommandTest(CommandTest):
 
     def test_arguments(self):
         # only pass one argument. spider script shouldn't be created
-        self.assertEqual(0, self.call('genspider', 'test_name'))
+        self.assertEqual(2, self.call('genspider', 'test_name'))
         assert not exists(join(self.proj_mod_path, 'spiders', 'test_name.py'))
         # pass two arguments <name> <domain>. spider script should be created
         self.assertEqual(0, self.call('genspider', 'test_name', 'test.com'))
@@ -147,12 +147,12 @@ from scrapy.spider import BaseSpider
 """)
         p = self.proc('runspider', fname)
         log = p.stderr.read()
-        self.assert_("ERROR: No spider found in file" in log)
+        self.assert_("No spider found in file" in log)
 
     def test_runspider_file_not_found(self):
         p = self.proc('runspider', 'some_non_existent_file')
         log = p.stderr.read()
-        self.assert_("ERROR: File not found: some_non_existent_file" in log)
+        self.assert_("File not found: some_non_existent_file" in log)
 
     def test_runspider_unable_to_load(self):
         tmpdir = self.mktemp()
@@ -162,5 +162,5 @@ from scrapy.spider import BaseSpider
             f.write("")
         p = self.proc('runspider', fname)
         log = p.stderr.read()
-        self.assert_("ERROR: Unable to load" in log)
+        self.assert_("Unable to load" in log)