diff --git a/visualdataset/__init__.py b/visualdataset/__init__.py
index 8423dad..2dd9179 100644
--- a/visualdataset/__init__.py
+++ b/visualdataset/__init__.py
@@ -9,4 +9,4 @@
 |_|
 """
 
-__version__ = '0.2.3'
+__version__ = '0.3.0'
diff --git a/visualdataset/__main__.py b/visualdataset/__main__.py
index e4dcca0..a77bad4 100644
--- a/visualdataset/__main__.py
+++ b/visualdataset/__main__.py
@@ -14,14 +14,14 @@
 parser = ArgumentParser(description='Prepares a dataset for use with the ChRIS_ui '
                                     '"Visual Datasets" feature.',
                         formatter_class=ArgumentDefaultsHelpFormatter)
-parser.add_argument('--mode', type=str, default='file', choices=['file', 'string', 'freesurfer-7.3.3', 'malpem-1.3'],
-                    help='File matching and option selection mode. file=accept JSON files from '
-                         '--matchers and --options. string=accept JSON strings from --matchers and --options. '
-                         'freesurfer=built-in support for the FreeSurfer output files.')
+parser.add_argument('--mode', type=str, default='none', choices=['freesurfer-7.3.3', 'malpem-1.3', 'none'],
+                    help='Use default values for --matchers and --options for a well known program\'s output. ')
 parser.add_argument('--matchers', type=str,
-                    help='Regular expressions used to assign tags to files')
+                    help='Regular expressions used to assign tags to files, '
+                         'as stringified JSON or a path to a file relative to inputdir')
 parser.add_argument('--options', type=str,
-                    help='Metadata to go with tag sets')
+                    help='Metadata to go with tag sets, '
+                         'as stringified JSON or a path to a file relative to inputdir')
 parser.add_argument('--first-run-files', type=str, default='[]',
                     help='List of files to show on first run, '
                          'as a stringified JSON list of paths relative to inputdir')
diff --git a/visualdataset/json_arg_parser.py b/visualdataset/json_arg_parser.py
index c462df7..2b5e7e6 100644
--- a/visualdataset/json_arg_parser.py
+++ b/visualdataset/json_arg_parser.py
@@ -10,30 +10,50 @@
 from visualdataset.wellknown import FREESURFER_MATCHERS, FREESURFER_OPTIONS, MALPEM_MATCHERS, MALPEM_OPTIONS
 
 
+EVERYTHING_MATCHER: Sequence[Matcher] = [Matcher(key='type', value='file', regex=r'\.(nii(\.gz)?)|(mgz)$')]
+"""
+a Matcher matching all supported file formats.
+"""
+
+
 def parse_args(input_dir: Path, mode: str, matchers: str | None, options: str | None
                ) -> tuple[Sequence[Matcher], Sequence[OptionsLink]]:
     mode = mode.lower()
-    if mode.startswith('freesurfer'):
-        return FREESURFER_MATCHERS, FREESURFER_OPTIONS
-    if mode.startswith('malpem'):
-        return MALPEM_MATCHERS, MALPEM_OPTIONS
-    if mode == 'file':
-        matchers_str = '[]' if matchers is None else (input_dir / matchers).read_text()
-        options_str = '[]' if options is None else (input_dir / options).read_text()
-    elif mode == 'string':
-        matchers_str = '[]' if matchers is None else matchers
-        options_str = '[]' if options is None else options
-    else:
-        print(f'Unsupported option --mode={mode}')
-        sys.exit(1)
-    matchers_list = deserialize_list(matchers_str, Matcher, '--matchers')
-    options_list = deserialize_list(options_str, OptionsLink, '--options')
+    matchers_list = None
+    options_list = []
+
+    if 'freesurfer' in mode:
+        matchers_list = FREESURFER_MATCHERS
+        options_list = FREESURFER_OPTIONS
+    if 'malpem' in mode:
+        matchers_list = MALPEM_MATCHERS
+        options_list = MALPEM_OPTIONS
+
+    if matchers is not None:
+        matchers_list = deserialize_list(file_or_string(input_dir, matchers), Matcher, '--matchers')
+    if options is not None:
+        options_list = deserialize_list(file_or_string(input_dir, options), OptionsLink, '--options')
+
+    if matchers_list is None:
+        matchers_list = EVERYTHING_MATCHER
+
     return matchers_list, options_list
 
 
 _M = TypeVar('_M', bound=BaseModel)
 
 
+def file_or_string(dir: Path, arg: str) -> str:
+    path = dir / arg
+    try:
+        is_file = path.is_file()
+    except OSError:  # in case file name is too long (it's probably the value!)
+        is_file = False
+    if is_file:
+        return path.read_text()
+    return arg
+
+
 def deserialize_list(s: str, t: Type[_M], flag: str) -> Sequence[_M]:
     try:
         data = json.loads(s)