diff --git a/TODO.txt b/TODO.txt
index f8b1934..d36a81f 100644
--- a/TODO.txt
+++ b/TODO.txt
@@ -1,3 +1,6 @@
+Add a convenient way to run parsers
+Add repetition parsers to avoid some recursion
+Think about not requiring so much Pos() when making messages
 Rename Seq2 to Seq
 Document Seq
 Add SPDX tags
diff --git a/bytes/regexp.go b/bytes/regexp.go
index dd3d7d6..aec7c53 100644
--- a/bytes/regexp.go
+++ b/bytes/regexp.go
@@ -5,7 +5,6 @@ import (
 	"fmt"
 	"io"
 	"regexp"
-	"strings"
 	"unicode/utf8"
 
 	"git.codemonkeysoftware.net/b/gigaparsec"
@@ -56,12 +55,10 @@ func (rr *RuneReader) Count() uint64 {
 	return rr.cursor.Pos() - rr.start
 }
 
-func Regexp(str string) gigaparsec.Parser[byte, string] {
-	if !strings.HasPrefix(str, "^") && !strings.HasPrefix(str, `\A`) {
-		str = "^" + str
-	}
-	re := regexp.MustCompile(str)
-	expected := fmt.Sprintf("match `%s`", str)
+func Regexp(pattern string) gigaparsec.Parser[byte, string] {
+	pattern = fmt.Sprintf(`^(?:%s)`, pattern)
+	re := regexp.MustCompile(pattern)
+	expected := fmt.Sprintf("match `%s`", pattern)
 	return func(input gigaparsec.State[byte]) (gigaparsec.Result[byte, string], error) {
 		r := NewRuneReader(input.Cursor())
 		idx := re.FindReaderIndex(r)