diff --git a/tests/rules/test_quoted_strings.py b/tests/rules/test_quoted_strings.py
index 543cc0d1..02e63cfa 100644
--- a/tests/rules/test_quoted_strings.py
+++ b/tests/rules/test_quoted_strings.py
@@ -258,7 +258,11 @@ def test_only_when_needed(self):
                    '  - foo\n'
                    '  - "foo"\n'                             # fails
                    'flow-seq: [foo, "foo"]\n'                # fails
-                   'flow-map: {a: foo, b: "foo"}\n',         # fails
+                   'flow-map: {a: foo, b: "foo"}\n'          # fails
+                   'flow-seq2: [foo, "foo,bar", "foo[bar]", "foo{bar}"]\n'
+                   'flow-map2: {a: foo, b: "foo,bar"}\n'
+                   'nested-flow1: {a: foo, b: [foo, "foo,bar"]}\n'
+                   'nested-flow2: [{a: foo}, {b: "foo,bar", c: ["foo[bar]"]}]\n',
                    conf, problem1=(5, 10), problem2=(8, 10), problem3=(18, 5),
                    problem4=(19, 17), problem5=(20, 23))
         self.check('---\n'
@@ -299,10 +303,12 @@ def test_only_when_needed_single_quotes(self):
                    '  - foo\n'
                    '  - "foo"\n'                             # fails
                    'flow-seq: [foo, "foo"]\n'                # fails
-                   'flow-map: {a: foo, b: "foo"}\n',         # fails
+                   'flow-map: {a: foo, b: "foo"}\n'          # fails
+                   'flow-seq2: [foo, "foo,bar"]\n'           # fails
+                   'flow-map2: {a: foo, b: "foo,bar"}\n',    # fails
                    conf, problem1=(5, 10), problem2=(6, 10), problem3=(7, 10),
                    problem4=(8, 10), problem5=(18, 5), problem6=(19, 17),
-                   problem7=(20, 23))
+                   problem7=(20, 23), problem8=(21, 18), problem9=(22, 24))
         self.check('---\n'
                    'multiline string 1: |\n'
                    '  line 1\n'
diff --git a/yamllint/rules/quoted_strings.py b/yamllint/rules/quoted_strings.py
index 9380ae57..86b825e0 100644
--- a/yamllint/rules/quoted_strings.py
+++ b/yamllint/rules/quoted_strings.py
@@ -186,7 +186,11 @@ def _quote_match(quote_type, token_style):
             (quote_type == 'double' and token_style == '"'))
 
 
-def _quotes_are_needed(string):
+def _quotes_are_needed(string, is_inside_a_flow):
+    # Quotes needed on strings containing flow tokens
+    if is_inside_a_flow and set(string) & {',', '[', ']', '{', '}'}:
+        return True
+
     loader = yaml.BaseLoader('key: ' + string)
     # Remove the 5 first tokens corresponding to 'key: ' (StreamStartToken,
     # BlockMappingStartToken, KeyToken, ScalarToken(value=key), ValueToken)
@@ -209,6 +213,16 @@ def _has_quoted_quotes(token):
 
 
 def check(conf, token, prev, next, nextnext, context):
+    if 'flow_nest_count' not in context:
+        context['flow_nest_count'] = 0
+
+    if isinstance(token, (yaml.FlowMappingStartToken,
+                          yaml.FlowSequenceStartToken)):
+        context['flow_nest_count'] += 1
+    elif isinstance(token, (yaml.FlowMappingEndToken,
+                            yaml.FlowSequenceEndToken)):
+        context['flow_nest_count'] -= 1
+
     if not (isinstance(token, yaml.tokens.ScalarToken) and
             isinstance(prev, (yaml.BlockEntryToken, yaml.FlowEntryToken,
                               yaml.FlowSequenceStartToken, yaml.TagToken,
@@ -261,7 +275,8 @@ def check(conf, token, prev, next, nextnext, context):
 
         # Quotes are not strictly needed here
         if (token.style and tag == DEFAULT_SCALAR_TAG and token.value and
-                not _quotes_are_needed(token.value)):
+                not _quotes_are_needed(token.value,
+                                       context['flow_nest_count'] > 0)):
             is_extra_required = any(re.search(r, token.value)
                                     for r in conf['extra-required'])
             is_extra_allowed = any(re.search(r, token.value)