diff --git a/codec/cbor.go b/codec/cbor.go
index 10944487..d6834f3b 100644
--- a/codec/cbor.go
+++ b/codec/cbor.go
@@ -427,12 +427,13 @@ func (d *cborDecDriver) decLen() int {
 	return int(d.decUint())
 }
 
-func (d *cborDecDriver) decAppendIndefiniteBytes(bs []byte) []byte {
+func (d *cborDecDriver) decAppendIndefiniteBytes(bs []byte, major byte) []byte {
 	d.bdRead = false
 	for !d.CheckBreak() {
-		if major := d.bd >> 5; major != cborMajorBytes && major != cborMajorString {
-			d.d.errorf("invalid indefinite string/bytes %x (%s); got major %v, expected %v or %v",
-				d.bd, cbordesc(d.bd), major, cborMajorBytes, cborMajorString)
+		chunkMajor := d.bd >> 5
+		if chunkMajor != major {
+			d.d.errorf("malformed indefinite string/bytes %x (%s); contains chunk with major type %v, expected %v",
+				d.bd, cbordesc(d.bd), chunkMajor, major)
 		}
 		n := uint(d.decLen())
 		oldLen := uint(len(bs))
@@ -445,6 +446,9 @@ func (d *cborDecDriver) decAppendIndefiniteBytes(bs []byte) []byte {
 			bs = bs[:newLen]
 		}
 		d.d.decRd.readb(bs[oldLen:newLen])
+		if d.h.ValidateUnicode && major == cborMajorString && !utf8.Valid(bs[oldLen:newLen]) {
+			d.d.errorf("indefinite-length text string contains chunk that is not a valid utf-8 sequence: 0x%x", bs[oldLen:newLen])
+		}
 		d.bdRead = false
 	}
 	d.bdRead = false
@@ -580,9 +584,9 @@ func (d *cborDecDriver) DecodeBytes(bs []byte) (bsOut []byte) {
 		d.bdRead = false
 		if bs == nil {
 			d.d.decByteState = decByteStateReuseBuf
-			return d.decAppendIndefiniteBytes(d.d.b[:0])
+			return d.decAppendIndefiniteBytes(d.d.b[:0], d.bd>>5)
 		}
-		return d.decAppendIndefiniteBytes(bs[:0])
+		return d.decAppendIndefiniteBytes(bs[:0], d.bd>>5)
 	}
 	if d.bd == cborBdIndefiniteArray {
 		d.bdRead = false
diff --git a/codec/cbor_test.go b/codec/cbor_test.go
index 9d62f4b9..6682bd62 100644
--- a/codec/cbor_test.go
+++ b/codec/cbor_test.go
@@ -61,8 +61,8 @@ func TestCborIndefiniteLength(t *testing.T) {
 	buf.WriteByte(cborBdBreak)
 
 	buf.WriteByte(cborBdIndefiniteString)
-	e.MustEncode([]byte("two-")) // encode as bytes, to check robustness of code
-	e.MustEncode([]byte("value"))
+	e.MustEncode("two-")
+	e.MustEncode("value")
 	buf.WriteByte(cborBdBreak)
 
 	//----
@@ -92,6 +92,41 @@ func TestCborIndefiniteLength(t *testing.T) {
 	}
 }
 
+// "If any item between the indefinite-length string indicator (0b010_11111 or 0b011_11111) and the
+// "break" stop code is not a definite-length string item of the same major type, the string is not
+// well-formed."
+func TestCborIndefiniteLengthStringChunksCannotMixTypes(t *testing.T) {
+	defer testSetup(t, nil)()
+	var handle CborHandle
+
+	for _, in := range [][]byte{
+		{cborBdIndefiniteString, 0x40, cborBdBreak}, // byte string chunk in indefinite length text string
+		{cborBdIndefiniteBytes, 0x60, cborBdBreak},  // text string chunk in indefinite length byte string
+	} {
+		var out string
+		err := NewDecoderBytes(in, &handle).Decode(&out)
+		if err == nil {
+			t.Errorf("expected error but decoded 0x%x to: %q", in, out)
+		}
+	}
+}
+
+// "If any definite-length text string inside an indefinite-length text string is invalid, the
+// indefinite-length text string is invalid. Note that this implies that the UTF-8 bytes of a single
+// Unicode code point (scalar value) cannot be spread between chunks: a new chunk of a text string
+// can only be started at a code point boundary."
+func TestCborIndefiniteLengthTextStringChunksAreUTF8(t *testing.T) {
+	defer testSetup(t, nil)()
+	var handle CborHandle
+	handle.ValidateUnicode = true
+
+	var out string
+	err := NewDecoderBytes([]byte{cborBdIndefiniteString, 0x61, 0xc2, 0x61, 0xa3, cborBdBreak}, &handle).Decode(&out)
+	if err == nil {
+		t.Errorf("expected error but decoded to: %q", out)
+	}
+}
+
 type testCborGolden struct {
 	Base64     string      `codec:"cbor"`
 	Hex        string      `codec:"hex"`