diff --git a/ast/ast.go b/ast/ast.go
index 778fbc0..5dc2383 100644
--- a/ast/ast.go
+++ b/ast/ast.go
@@ -690,10 +690,15 @@ func (f *ForOf) Children() []Node {
 
 // Children returns the node's child nodes.
 func (o *Of) Children() []Node {
-	nodes := []Node{o.Quantifier, o.Strings}
+	// Because this node can have children that are exclusively rules or
+	// strings we need to only add them if they are non-nil.
+	nodes := []Node{o.Quantifier}
 	if o.Rules != nil {
 		nodes = append(nodes, o.Rules)
 	}
+	if o.Strings != nil {
+		nodes = append(nodes, o.Strings)
+	}
 	return nodes
 }
 
@@ -1233,5 +1238,3 @@ func (o *Operation) AsProto() *pb.Expression {
 	}
 	return expr
 }
-
-
diff --git a/error/error.go b/error/error.go
index 8047a17..3e6841d 100644
--- a/error/error.go
+++ b/error/error.go
@@ -30,6 +30,7 @@ const (
 	UnevenNumberOfDigitsError
 	InvalidAsciiError
 	InvalidUTF8Error
+	UndefinedStringIdentifierError
 )
 
 type Error struct {
diff --git a/parser/adapter.go b/parser/adapter.go
index 342d5fc..f71a442 100644
--- a/parser/adapter.go
+++ b/parser/adapter.go
@@ -32,6 +32,10 @@ func Parse(input io.Reader) (rs *ast.RuleSet, err error) {
 			Imports: make([]string, 0),
 			Rules:   make([]*ast.Rule, 0),
 		},
+		// Used to collect the strings as they are parsed on a per-rule basis.
+		// When the condition is parsed this is used as a lookup table to check
+		// for undefined strings.
+		strings: make(map[string]bool),
 	}
 	lexer.scanner.In = input
 	lexer.scanner.Out = ioutil.Discard
@@ -51,6 +55,7 @@ type lexer struct {
 	scanner Scanner
 	err     gyperror.Error
 	ruleSet *ast.RuleSet
+	strings map[string]bool
 }
 
 // Lex provides the interface expected by the goyacc parser. This function is
diff --git a/parser/grammar.y b/parser/grammar.y
index e6b190c..0de5929 100644
--- a/parser/grammar.y
+++ b/parser/grammar.y
@@ -314,6 +314,9 @@ rule
       {
         $<rule>4.Condition = $10
         $$ = $<rule>4
+
+        // Clear the strings map for the next rule being parsed.
+        asLexer(yrlex).strings = make(map[string]bool)
       }
     ;
 
@@ -470,10 +473,14 @@ meta_declaration
 string_declarations
     : string_declaration
       {
+        lexer := asLexer(yrlex)
+        lexer.strings[$1.GetIdentifier()] = true
         $$ = []ast.String{$1}
       }
     | string_declarations string_declaration
       {
+        lexer := asLexer(yrlex)
+        lexer.strings[$2.GetIdentifier()] = true
         $$ = append($1, $2)
       }
     ;
@@ -829,21 +836,51 @@ expression
       }
     | _STRING_IDENTIFIER_
       {
+        identifier := strings.TrimPrefix($1, "$")
+        // Exclude anonymous ($) strings.
+        if identifier != "" {
+          lexer := asLexer(yrlex)
+          if _, ok := lexer.strings[identifier]; !ok {
+            return lexer.setError(
+              gyperror.UndefinedStringIdentifierError,
+              `undefined string identifier: %s`, $1)
+          }
+        }
         $$ = &ast.StringIdentifier{
-          Identifier: strings.TrimPrefix($1, "$"),
+          Identifier: identifier,
         }
       }
     | _STRING_IDENTIFIER_ _AT_ primary_expression
       {
+        identifier := strings.TrimPrefix($1, "$")
+        // Exclude anonymous ($) strings.
+        if identifier != "" {
+          lexer := asLexer(yrlex)
+          if _, ok := lexer.strings[identifier]; !ok {
+            return lexer.setError(
+              gyperror.UndefinedStringIdentifierError,
+              `undefined string identifier: %s`, $1)
+          }
+        }
         $$ = &ast.StringIdentifier{
-          Identifier: strings.TrimPrefix($1, "$"),
+          Identifier: identifier,
           At: $3,
         }
       }
     | _STRING_IDENTIFIER_ _IN_ range
       {
+        identifier := strings.TrimPrefix($1, "$")
+        // Exclude anonymous ($) strings.
+        if identifier != "" {
+          lexer := asLexer(yrlex)
+          if _, ok := lexer.strings[identifier]; !ok {
+            return lexer.setError(
+              gyperror.UndefinedStringIdentifierError,
+              `undefined string identifier: %s`, $1)
+          }
+        }
         $$ = &ast.StringIdentifier{
-          Identifier: strings.TrimPrefix($1, "$"),
+          Identifier: identifier,
           In: $3,
         }
       }
@@ -1011,6 +1048,12 @@ string_set
       }
     | _THEM_
       {
+        lexer := asLexer(yrlex)
+        if len(lexer.strings) == 0 {
+          return lexer.setError(
+            gyperror.UndefinedStringIdentifierError,
+            `undefined string identifier: %s`, ast.KeywordThem)
+        }
         $$ = ast.KeywordThem
       }
     ;
@@ -1031,16 +1074,48 @@ string_enumeration
 string_enumeration_item
     : _STRING_IDENTIFIER_
       {
+        identifier := strings.TrimPrefix($1, "$")
+        lexer := asLexer(yrlex)
+        // Anonymous strings ($) in string enumerations are an error.
+        if _, ok := lexer.strings[identifier]; !ok || identifier == "" {
+          return lexer.setError(
+            gyperror.UndefinedStringIdentifierError,
+            `undefined string identifier: %s`, $1)
+        }
         $$ = &ast.StringIdentifier{
-          Identifier: strings.TrimPrefix($1, "$"),
+          Identifier: identifier,
         }
       }
     | _STRING_IDENTIFIER_WITH_WILDCARD_
-      {
-        $$ = &ast.StringIdentifier{
-          Identifier: strings.TrimPrefix($1, "$"),
+    {
+      identifier := strings.TrimSuffix($1, "*")
+      lexer := asLexer(yrlex)
+      // There must be at least one defined string.
+      if len(identifier) == 0 && len(lexer.strings) == 0 {
+          return lexer.setError(
+            gyperror.UndefinedStringIdentifierError,
+            `undefined string identifier: %s`, $1)
+      }
+
+      // There must be at least one string that will match the wildcard.
+      identifier = strings.TrimPrefix(identifier, "$")
+      match := false
+      for s, _ := range lexer.strings {
+        if strings.HasPrefix(s, identifier) {
+          match = true
+          break
         }
       }
+      if !match {
+        return lexer.setError(
+          gyperror.UndefinedStringIdentifierError,
+          `undefined string identifier: %s`, $1)
+      }
+      // Can't use "identifier" here as that has the asterisk stripped already.
+      $$ = &ast.StringIdentifier{
+        Identifier: strings.TrimPrefix($1, "$"),
+      }
+    }
     ;
 
 
@@ -1158,39 +1233,93 @@ primary_expression
       }
     | _STRING_COUNT_ _IN_ range
       {
+        identifier := strings.TrimPrefix($1, "#")
+        if identifier != "" {
+          lexer := asLexer(yrlex)
+          if _, ok := lexer.strings[identifier]; !ok {
+            return lexer.setError(
+              gyperror.UndefinedStringIdentifierError,
+              `undefined string identifier: %s`, $1)
+          }
+        }
         $$ = &ast.StringCount{
-          Identifier: strings.TrimPrefix($1, "#"),
+          Identifier: identifier,
           In: $3,
         }
       }
     | _STRING_COUNT_
       {
+        identifier := strings.TrimPrefix($1, "#")
+        if identifier != "" {
+          lexer := asLexer(yrlex)
+          if _, ok := lexer.strings[identifier]; !ok {
+            return lexer.setError(
+              gyperror.UndefinedStringIdentifierError,
+              `undefined string identifier: %s`, $1)
+          }
+        }
         $$ = &ast.StringCount{
-          Identifier: strings.TrimPrefix($1, "#"),
+          Identifier: identifier,
         }
       }
     | _STRING_OFFSET_ '[' primary_expression ']'
       {
+        identifier := strings.TrimPrefix($1, "@")
+        if identifier != "" {
+          lexer := asLexer(yrlex)
+          if _, ok := lexer.strings[identifier]; !ok {
+            return lexer.setError(
+              gyperror.UndefinedStringIdentifierError,
+              `undefined string identifier: %s`, $1)
+          }
+        }
         $$ = &ast.StringOffset{
-          Identifier: strings.TrimPrefix($1, "@"),
+          Identifier: identifier,
           Index: $3,
         }
       }
     | _STRING_OFFSET_
       {
+        identifier := strings.TrimPrefix($1, "@")
+        if identifier != "" {
+          lexer := asLexer(yrlex)
+          if _, ok := lexer.strings[identifier]; !ok {
+            return lexer.setError(
+              gyperror.UndefinedStringIdentifierError,
+              `undefined string identifier: %s`, $1)
+          }
+        }
         $$ = &ast.StringOffset{
-          Identifier: strings.TrimPrefix($1, "@"),
+          Identifier: identifier,
         }
       }
     | _STRING_LENGTH_ '[' primary_expression ']'
       {
+        identifier := strings.TrimPrefix($1, "!")
+        if identifier != "" {
+          lexer := asLexer(yrlex)
+          if _, ok := lexer.strings[identifier]; !ok {
+            return lexer.setError(
+              gyperror.UndefinedStringIdentifierError,
+              `undefined string identifier: %s`, $1)
+          }
+        }
         $$ = &ast.StringLength{
-          Identifier: strings.TrimPrefix($1, "!"),
+          Identifier: identifier,
           Index: $3,
         }
       }
     | _STRING_LENGTH_
       {
+        identifier := strings.TrimPrefix($1, "!")
+        if identifier != "" {
+          lexer := asLexer(yrlex)
+          if _, ok := lexer.strings[identifier]; !ok {
+            return lexer.setError(
+              gyperror.UndefinedStringIdentifierError,
+              `undefined string identifier: %s`, $1)
+          }
+        }
         $$ = &ast.StringLength{
           Identifier: strings.TrimPrefix($1, "!"),
         }
@@ -1272,4 +1401,4 @@ func operation(operator ast.OperatorType, left, right ast.Expression) (n ast.Exp
     }
   }
   return n
-}
+}
\ No newline at end of file
diff --git a/parser/parser.go b/parser/parser.go
index 329f9c9..5936ed3 100644
--- a/parser/parser.go
+++ b/parser/parser.go
@@ -237,7 +237,7 @@ const yrEofCode = 1
 const yrErrCode = 2
 const yrInitialStackSize = 16
 
-//line parser/grammar.y:1256
+//line parser/grammar.y:1385
 
 // This function takes an operator and two operands and returns a Expression
 // representing the operation. If the left operand is an operation of the
@@ -936,47 +936,50 @@ yrdefault:
 		{
 			yrDollar[4].rule.Condition = yrDollar[10].expr
 			yrVAL.rule = yrDollar[4].rule
+
+			// Clear the strings map for the next rule being parsed.
+			asLexer(yrlex).strings = make(map[string]bool)
 		}
 	case 10:
 		yrDollar = yrS[yrpt-0 : yrpt+1]
-//line parser/grammar.y:323
+//line parser/grammar.y:326
 		{
 			yrVAL.metas = []*ast.Meta{}
 		}
 	case 11:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:327
+//line parser/grammar.y:330
 		{
 			yrVAL.metas = yrDollar[3].metas
 		}
 	case 12:
 		yrDollar = yrS[yrpt-0 : yrpt+1]
-//line parser/grammar.y:335
+//line parser/grammar.y:338
 		{
 			yrVAL.yss = []ast.String{}
 		}
 	case 13:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:339
+//line parser/grammar.y:342
 		{
 			yrVAL.yss = yrDollar[3].yss
 		}
 	case 14:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:347
+//line parser/grammar.y:350
 		{
 			yrVAL.expr = yrDollar[3].expr
 		}
 	case 15:
 		yrDollar = yrS[yrpt-0 : yrpt+1]
-//line parser/grammar.y:355
+//line parser/grammar.y:358
 		{
 			yrVAL.mod = 0
 			yrVAL.lineno = -1
 		}
 	case 16:
 		yrDollar = yrS[yrpt-2 : yrpt+1]
-//line parser/grammar.y:360
+//line parser/grammar.y:363
 		{
 			yrVAL.mod = yrDollar[1].mod | yrDollar[2].mod
 
@@ -988,39 +991,39 @@ yrdefault:
 		}
 	case 17:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:374
+//line parser/grammar.y:377
 		{
 			yrVAL.mod = ModPrivate
 			yrVAL.lineno = yrDollar[1].lineno
 		}
 	case 18:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:379
+//line parser/grammar.y:382
 		{
 			yrVAL.mod = ModGlobal
 			yrVAL.lineno = yrDollar[1].lineno
 		}
 	case 19:
 		yrDollar = yrS[yrpt-0 : yrpt+1]
-//line parser/grammar.y:388
+//line parser/grammar.y:391
 		{
 			yrVAL.ss = []string{}
 		}
 	case 20:
 		yrDollar = yrS[yrpt-2 : yrpt+1]
-//line parser/grammar.y:392
+//line parser/grammar.y:395
 		{
 			yrVAL.ss = yrDollar[2].ss
 		}
 	case 21:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:400
+//line parser/grammar.y:403
 		{
 			yrVAL.ss = []string{yrDollar[1].s}
 		}
 	case 22:
 		yrDollar = yrS[yrpt-2 : yrpt+1]
-//line parser/grammar.y:404
+//line parser/grammar.y:407
 		{
 			lexer := asLexer(yrlex)
 
@@ -1035,19 +1038,19 @@ yrdefault:
 		}
 	case 23:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:421
+//line parser/grammar.y:424
 		{
 			yrVAL.metas = []*ast.Meta{yrDollar[1].meta}
 		}
 	case 24:
 		yrDollar = yrS[yrpt-2 : yrpt+1]
-//line parser/grammar.y:425
+//line parser/grammar.y:428
 		{
 			yrVAL.metas = append(yrDollar[1].metas, yrDollar[2].meta)
 		}
 	case 25:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:433
+//line parser/grammar.y:436
 		{
 			yrVAL.meta = &ast.Meta{
 				Key:   yrDollar[1].s,
@@ -1056,7 +1059,7 @@ yrdefault:
 		}
 	case 26:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:440
+//line parser/grammar.y:443
 		{
 			yrVAL.meta = &ast.Meta{
 				Key:   yrDollar[1].s,
@@ -1065,7 +1068,7 @@ yrdefault:
 		}
 	case 27:
 		yrDollar = yrS[yrpt-4 : yrpt+1]
-//line parser/grammar.y:447
+//line parser/grammar.y:450
 		{
 			yrVAL.meta = &ast.Meta{
 				Key:   yrDollar[1].s,
@@ -1074,7 +1077,7 @@ yrdefault:
 		}
 	case 28:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:454
+//line parser/grammar.y:457
 		{
 			yrVAL.meta = &ast.Meta{
 				Key:   yrDollar[1].s,
@@ -1083,7 +1086,7 @@ yrdefault:
 		}
 	case 29:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:461
+//line parser/grammar.y:464
 		{
 			yrVAL.meta = &ast.Meta{
 				Key:   yrDollar[1].s,
@@ -1092,19 +1095,23 @@ yrdefault:
 		}
 	case 30:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:472
+//line parser/grammar.y:475
 		{
+			lexer := asLexer(yrlex)
+			lexer.strings[yrDollar[1].ys.GetIdentifier()] = true
 			yrVAL.yss = []ast.String{yrDollar[1].ys}
 		}
 	case 31:
 		yrDollar = yrS[yrpt-2 : yrpt+1]
-//line parser/grammar.y:476
+//line parser/grammar.y:481
 		{
+			lexer := asLexer(yrlex)
+			lexer.strings[yrDollar[2].ys.GetIdentifier()] = true
 			yrVAL.yss = append(yrDollar[1].yss, yrDollar[2].ys)
 		}
 	case 32:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:484
+//line parser/grammar.y:491
 		{
 			if err := validateUTF8(yrDollar[3].s); err != nil {
 				return asLexer(yrlex).setError(
@@ -1113,7 +1120,7 @@ yrdefault:
 		}
 	case 33:
 		yrDollar = yrS[yrpt-5 : yrpt+1]
-//line parser/grammar.y:491
+//line parser/grammar.y:498
 		{
 			yrVAL.ys = &ast.TextString{
 				BaseString: ast.BaseString{
@@ -1136,7 +1143,7 @@ yrdefault:
 		}
 	case 34:
 		yrDollar = yrS[yrpt-4 : yrpt+1]
-//line parser/grammar.y:512
+//line parser/grammar.y:519
 		{
 			yrVAL.ys = &ast.RegexpString{
 				BaseString: ast.BaseString{
@@ -1153,7 +1160,7 @@ yrdefault:
 		}
 	case 35:
 		yrDollar = yrS[yrpt-4 : yrpt+1]
-//line parser/grammar.y:527
+//line parser/grammar.y:534
 		{
 			yrVAL.ys = &ast.HexString{
 				BaseString: ast.BaseString{
@@ -1166,13 +1173,13 @@ yrdefault:
 		}
 	case 36:
 		yrDollar = yrS[yrpt-0 : yrpt+1]
-//line parser/grammar.y:542
+//line parser/grammar.y:549
 		{
 			yrVAL.smod = stringModifiers{}
 		}
 	case 37:
 		yrDollar = yrS[yrpt-2 : yrpt+1]
-//line parser/grammar.y:546
+//line parser/grammar.y:553
 		{
 			if yrDollar[1].smod.modifiers&yrDollar[2].smod.modifiers != 0 {
 				return asLexer(yrlex).setError(
@@ -1194,49 +1201,49 @@ yrdefault:
 		}
 	case 38:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:569
+//line parser/grammar.y:576
 		{
 			yrVAL.smod = stringModifiers{modifiers: ModWide}
 		}
 	case 39:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:570
+//line parser/grammar.y:577
 		{
 			yrVAL.smod = stringModifiers{modifiers: ModASCII}
 		}
 	case 40:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:571
+//line parser/grammar.y:578
 		{
 			yrVAL.smod = stringModifiers{modifiers: ModNocase}
 		}
 	case 41:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:572
+//line parser/grammar.y:579
 		{
 			yrVAL.smod = stringModifiers{modifiers: ModFullword}
 		}
 	case 42:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:573
+//line parser/grammar.y:580
 		{
 			yrVAL.smod = stringModifiers{modifiers: ModPrivate}
 		}
 	case 43:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:574
+//line parser/grammar.y:581
 		{
 			yrVAL.smod = stringModifiers{modifiers: ModBase64}
 		}
 	case 44:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:575
+//line parser/grammar.y:582
 		{
 			yrVAL.smod = stringModifiers{modifiers: ModBase64Wide}
 		}
 	case 45:
 		yrDollar = yrS[yrpt-4 : yrpt+1]
-//line parser/grammar.y:577
+//line parser/grammar.y:584
 		{
 			if err := validateAscii(yrDollar[3].s); err != nil {
 				return asLexer(yrlex).setError(
@@ -1256,7 +1263,7 @@ yrdefault:
 		}
 	case 46:
 		yrDollar = yrS[yrpt-4 : yrpt+1]
-//line parser/grammar.y:595
+//line parser/grammar.y:602
 		{
 			if err := validateAscii(yrDollar[3].s); err != nil {
 				return asLexer(yrlex).setError(
@@ -1276,7 +1283,7 @@ yrdefault:
 		}
 	case 47:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:613
+//line parser/grammar.y:620
 		{
 			yrVAL.smod = stringModifiers{
 				modifiers: ModXor,
@@ -1286,7 +1293,7 @@ yrdefault:
 		}
 	case 48:
 		yrDollar = yrS[yrpt-4 : yrpt+1]
-//line parser/grammar.y:621
+//line parser/grammar.y:628
 		{
 			yrVAL.smod = stringModifiers{
 				modifiers: ModXor,
@@ -1296,7 +1303,7 @@ yrdefault:
 		}
 	case 49:
 		yrDollar = yrS[yrpt-6 : yrpt+1]
-//line parser/grammar.y:629
+//line parser/grammar.y:636
 		{
 			lexer := asLexer(yrlex)
 
@@ -1326,73 +1333,73 @@ yrdefault:
 		}
 	case 50:
 		yrDollar = yrS[yrpt-0 : yrpt+1]
-//line parser/grammar.y:661
+//line parser/grammar.y:668
 		{
 			yrVAL.mod = 0
 		}
 	case 51:
 		yrDollar = yrS[yrpt-2 : yrpt+1]
-//line parser/grammar.y:665
+//line parser/grammar.y:672
 		{
 			yrVAL.mod = yrDollar[1].mod | yrDollar[2].mod
 		}
 	case 52:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:672
+//line parser/grammar.y:679
 		{
 			yrVAL.mod = ModWide
 		}
 	case 53:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:673
+//line parser/grammar.y:680
 		{
 			yrVAL.mod = ModASCII
 		}
 	case 54:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:674
+//line parser/grammar.y:681
 		{
 			yrVAL.mod = ModNocase
 		}
 	case 55:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:675
+//line parser/grammar.y:682
 		{
 			yrVAL.mod = ModFullword
 		}
 	case 56:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:676
+//line parser/grammar.y:683
 		{
 			yrVAL.mod = ModPrivate
 		}
 	case 57:
 		yrDollar = yrS[yrpt-0 : yrpt+1]
-//line parser/grammar.y:682
+//line parser/grammar.y:689
 		{
 			yrVAL.mod = 0
 		}
 	case 58:
 		yrDollar = yrS[yrpt-2 : yrpt+1]
-//line parser/grammar.y:686
+//line parser/grammar.y:693
 		{
 			yrVAL.mod = yrDollar[1].mod | yrDollar[2].mod
 		}
 	case 59:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:693
+//line parser/grammar.y:700
 		{
 			yrVAL.mod = ModPrivate
 		}
 	case 60:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:699
+//line parser/grammar.y:706
 		{
 			yrVAL.expr = &ast.Identifier{Identifier: yrDollar[1].s}
 		}
 	case 61:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:703
+//line parser/grammar.y:710
 		{
 			yrVAL.expr = &ast.MemberAccess{
 				Container: yrDollar[1].expr,
@@ -1401,7 +1408,7 @@ yrdefault:
 		}
 	case 62:
 		yrDollar = yrS[yrpt-4 : yrpt+1]
-//line parser/grammar.y:710
+//line parser/grammar.y:717
 		{
 			yrVAL.expr = &ast.Subscripting{
 				Array: yrDollar[1].expr,
@@ -1410,7 +1417,7 @@ yrdefault:
 		}
 	case 63:
 		yrDollar = yrS[yrpt-4 : yrpt+1]
-//line parser/grammar.y:717
+//line parser/grammar.y:724
 		{
 			yrVAL.expr = &ast.FunctionCall{
 				Callable:  yrDollar[1].expr,
@@ -1419,55 +1426,55 @@ yrdefault:
 		}
 	case 64:
 		yrDollar = yrS[yrpt-0 : yrpt+1]
-//line parser/grammar.y:728
+//line parser/grammar.y:735
 		{
 			yrVAL.exprs = []ast.Expression{}
 		}
 	case 65:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:732
+//line parser/grammar.y:739
 		{
 			yrVAL.exprs = yrDollar[1].exprs
 		}
 	case 66:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:739
+//line parser/grammar.y:746
 		{
 			yrVAL.exprs = []ast.Expression{yrDollar[1].expr}
 		}
 	case 67:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:743
+//line parser/grammar.y:750
 		{
 			yrVAL.exprs = append(yrDollar[1].exprs, yrDollar[3].expr)
 		}
 	case 68:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:751
+//line parser/grammar.y:758
 		{
 			yrVAL.reg = yrDollar[1].reg
 		}
 	case 69:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:759
+//line parser/grammar.y:766
 		{
 			yrVAL.expr = yrDollar[1].expr
 		}
 	case 70:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:767
+//line parser/grammar.y:774
 		{
 			yrVAL.expr = ast.KeywordTrue
 		}
 	case 71:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:771
+//line parser/grammar.y:778
 		{
 			yrVAL.expr = ast.KeywordFalse
 		}
 	case 72:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:775
+//line parser/grammar.y:782
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpMatches,
@@ -1476,7 +1483,7 @@ yrdefault:
 		}
 	case 73:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:782
+//line parser/grammar.y:789
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpContains,
@@ -1485,7 +1492,7 @@ yrdefault:
 		}
 	case 74:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:789
+//line parser/grammar.y:796
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpIContains,
@@ -1494,7 +1501,7 @@ yrdefault:
 		}
 	case 75:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:796
+//line parser/grammar.y:803
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpStartsWith,
@@ -1503,7 +1510,7 @@ yrdefault:
 		}
 	case 76:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:803
+//line parser/grammar.y:810
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpIStartsWith,
@@ -1512,7 +1519,7 @@ yrdefault:
 		}
 	case 77:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:810
+//line parser/grammar.y:817
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpEndsWith,
@@ -1521,7 +1528,7 @@ yrdefault:
 		}
 	case 78:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:817
+//line parser/grammar.y:824
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpIEndsWith,
@@ -1530,7 +1537,7 @@ yrdefault:
 		}
 	case 79:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:824
+//line parser/grammar.y:831
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpIEquals,
@@ -1539,33 +1546,63 @@ yrdefault:
 		}
 	case 80:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:831
+//line parser/grammar.y:838
 		{
+			identifier := strings.TrimPrefix(yrDollar[1].s, "$")
+			// Exclude anonymous ($) strings.
+			if identifier != "" {
+				lexer := asLexer(yrlex)
+				if _, ok := lexer.strings[identifier]; !ok {
+					return lexer.setError(
+						gyperror.UndefinedStringIdentifierError,
+						`undefined string identifier: %s`, yrDollar[1].s)
+				}
+			}
 			yrVAL.expr = &ast.StringIdentifier{
-				Identifier: strings.TrimPrefix(yrDollar[1].s, "$"),
+				Identifier: identifier,
 			}
 		}
 	case 81:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:837
+//line parser/grammar.y:854
 		{
+			identifier := strings.TrimPrefix(yrDollar[1].s, "$")
+			// Exclude anonymous ($) strings.
+			if identifier != "" {
+				lexer := asLexer(yrlex)
+				if _, ok := lexer.strings[identifier]; !ok {
+					return lexer.setError(
+						gyperror.UndefinedStringIdentifierError,
+						`undefined string identifier: %s`, yrDollar[1].s)
+				}
+			}
 			yrVAL.expr = &ast.StringIdentifier{
-				Identifier: strings.TrimPrefix(yrDollar[1].s, "$"),
+				Identifier: identifier,
 				At:         yrDollar[3].expr,
 			}
 		}
 	case 82:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:844
+//line parser/grammar.y:871
 		{
+			identifier := strings.TrimPrefix(yrDollar[1].s, "$")
+			// Exclude anonymous ($) strings.
+			if identifier != "" {
+				lexer := asLexer(yrlex)
+				if _, ok := lexer.strings[identifier]; !ok {
+					return lexer.setError(
+						gyperror.UndefinedStringIdentifierError,
+						`undefined string identifier: %s`, yrDollar[1].s)
+				}
+			}
 			yrVAL.expr = &ast.StringIdentifier{
-				Identifier: strings.TrimPrefix(yrDollar[1].s, "$"),
+				Identifier: identifier,
 				In:         yrDollar[3].rng,
 			}
 		}
 	case 83:
 		yrDollar = yrS[yrpt-9 : yrpt+1]
-//line parser/grammar.y:851
+//line parser/grammar.y:888
 		{
 			yrVAL.expr = &ast.ForIn{
 				Quantifier: yrDollar[2].expr,
@@ -1576,7 +1613,7 @@ yrdefault:
 		}
 	case 84:
 		yrDollar = yrS[yrpt-8 : yrpt+1]
-//line parser/grammar.y:860
+//line parser/grammar.y:897
 		{
 			yrVAL.expr = &ast.ForOf{
 				Quantifier: yrDollar[2].expr,
@@ -1586,7 +1623,7 @@ yrdefault:
 		}
 	case 85:
 		yrDollar = yrS[yrpt-5 : yrpt+1]
-//line parser/grammar.y:868
+//line parser/grammar.y:905
 		{
 			yrVAL.expr = &ast.Of{
 				Quantifier: yrDollar[1].expr,
@@ -1596,7 +1633,7 @@ yrdefault:
 		}
 	case 86:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:876
+//line parser/grammar.y:913
 		{
 			yrVAL.expr = &ast.Of{
 				Quantifier: yrDollar[1].expr,
@@ -1605,7 +1642,7 @@ yrdefault:
 		}
 	case 87:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:883
+//line parser/grammar.y:920
 		{
 			yrVAL.expr = &ast.Of{
 				Quantifier: yrDollar[1].expr,
@@ -1614,7 +1651,7 @@ yrdefault:
 		}
 	case 88:
 		yrDollar = yrS[yrpt-4 : yrpt+1]
-//line parser/grammar.y:890
+//line parser/grammar.y:927
 		{
 			yrVAL.expr = &ast.Of{
 				Quantifier: &ast.Percentage{yrDollar[1].expr},
@@ -1623,7 +1660,7 @@ yrdefault:
 		}
 	case 89:
 		yrDollar = yrS[yrpt-4 : yrpt+1]
-//line parser/grammar.y:897
+//line parser/grammar.y:934
 		{
 			yrVAL.expr = &ast.Of{
 				Quantifier: &ast.Percentage{yrDollar[1].expr},
@@ -1632,31 +1669,31 @@ yrdefault:
 		}
 	case 90:
 		yrDollar = yrS[yrpt-2 : yrpt+1]
-//line parser/grammar.y:904
+//line parser/grammar.y:941
 		{
 			yrVAL.expr = &ast.Not{yrDollar[2].expr}
 		}
 	case 91:
 		yrDollar = yrS[yrpt-2 : yrpt+1]
-//line parser/grammar.y:908
+//line parser/grammar.y:945
 		{
 			yrVAL.expr = &ast.Defined{yrDollar[2].expr}
 		}
 	case 92:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:912
+//line parser/grammar.y:949
 		{
 			yrVAL.expr = operation(ast.OpAnd, yrDollar[1].expr, yrDollar[3].expr)
 		}
 	case 93:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:916
+//line parser/grammar.y:953
 		{
 			yrVAL.expr = operation(ast.OpOr, yrDollar[1].expr, yrDollar[3].expr)
 		}
 	case 94:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:920
+//line parser/grammar.y:957
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpLessThan,
@@ -1665,7 +1702,7 @@ yrdefault:
 		}
 	case 95:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:927
+//line parser/grammar.y:964
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpGreaterThan,
@@ -1674,7 +1711,7 @@ yrdefault:
 		}
 	case 96:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:934
+//line parser/grammar.y:971
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpLessOrEqual,
@@ -1683,7 +1720,7 @@ yrdefault:
 		}
 	case 97:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:941
+//line parser/grammar.y:978
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpGreaterOrEqual,
@@ -1692,7 +1729,7 @@ yrdefault:
 		}
 	case 98:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:948
+//line parser/grammar.y:985
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpEqual,
@@ -1701,7 +1738,7 @@ yrdefault:
 		}
 	case 99:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:955
+//line parser/grammar.y:992
 		{
 			yrVAL.expr = &ast.Operation{
 				Operator: ast.OpNotEqual,
@@ -1710,31 +1747,31 @@ yrdefault:
 		}
 	case 100:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:962
+//line parser/grammar.y:999
 		{
 			yrVAL.expr = yrDollar[1].expr
 		}
 	case 101:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:966
+//line parser/grammar.y:1003
 		{
 			yrVAL.expr = &ast.Group{yrDollar[2].expr}
 		}
 	case 102:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:974
+//line parser/grammar.y:1011
 		{
 			yrVAL.node = &ast.Enum{Values: yrDollar[2].exprs}
 		}
 	case 103:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:978
+//line parser/grammar.y:1015
 		{
 			yrVAL.node = yrDollar[1].rng
 		}
 	case 104:
 		yrDollar = yrS[yrpt-5 : yrpt+1]
-//line parser/grammar.y:986
+//line parser/grammar.y:1023
 		{
 			yrVAL.rng = &ast.Range{
 				Start: yrDollar[2].expr,
@@ -1743,155 +1780,193 @@ yrdefault:
 		}
 	case 105:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:997
+//line parser/grammar.y:1034
 		{
 			yrVAL.exprs = []ast.Expression{yrDollar[1].expr}
 		}
 	case 106:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1001
+//line parser/grammar.y:1038
 		{
 			yrVAL.exprs = append(yrDollar[1].exprs, yrDollar[3].expr)
 		}
 	case 107:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1009
+//line parser/grammar.y:1046
 		{
 			yrVAL.node = &ast.Enum{Values: yrDollar[2].exprs}
 		}
 	case 108:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1013
+//line parser/grammar.y:1050
 		{
+			lexer := asLexer(yrlex)
+			if len(lexer.strings) == 0 {
+				return lexer.setError(
+					gyperror.UndefinedStringIdentifierError,
+					`undefined string identifier: %s`, ast.KeywordThem)
+			}
 			yrVAL.node = ast.KeywordThem
 		}
 	case 109:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1021
+//line parser/grammar.y:1064
 		{
 			yrVAL.exprs = []ast.Expression{yrDollar[1].si}
 		}
 	case 110:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1025
+//line parser/grammar.y:1068
 		{
 			yrVAL.exprs = append(yrDollar[1].exprs, yrDollar[3].si)
 		}
 	case 111:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1033
+//line parser/grammar.y:1076
 		{
+			identifier := strings.TrimPrefix(yrDollar[1].s, "$")
+			lexer := asLexer(yrlex)
+			// Anonymous strings ($) in string enumerations are an error.
+			if _, ok := lexer.strings[identifier]; !ok || identifier == "" {
+				return lexer.setError(
+					gyperror.UndefinedStringIdentifierError,
+					`undefined string identifier: %s`, yrDollar[1].s)
+			}
 			yrVAL.si = &ast.StringIdentifier{
-				Identifier: strings.TrimPrefix(yrDollar[1].s, "$"),
+				Identifier: identifier,
 			}
 		}
 	case 112:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1039
+//line parser/grammar.y:1090
 		{
+			identifier := strings.TrimSuffix(yrDollar[1].s, "*")
+			lexer := asLexer(yrlex)
+			// There must be at least one defined string.
+			if len(identifier) == 0 && len(lexer.strings) == 0 {
+				return lexer.setError(
+					gyperror.UndefinedStringIdentifierError,
+					`undefined string identifier: %s`, yrDollar[1].s)
+			}
+
+			// There must be at least one string that will match the wildcard.
+			identifier = strings.TrimPrefix(identifier, "$")
+			match := false
+			for s, _ := range lexer.strings {
+				if strings.HasPrefix(s, identifier) {
+					match = true
+					break
+				}
+			}
+			if !match {
+				return lexer.setError(
+					gyperror.UndefinedStringIdentifierError,
+					`undefined string identifier: %s`, yrDollar[1].s)
+			}
+			// Can't use "identifier" here as that has the asterisk stripped already.
 			yrVAL.si = &ast.StringIdentifier{
 				Identifier: strings.TrimPrefix(yrDollar[1].s, "$"),
 			}
 		}
 	case 113:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1049
+//line parser/grammar.y:1124
 		{
 			yrVAL.node = &ast.Enum{Values: yrDollar[2].exprs}
 		}
 	case 114:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1057
+//line parser/grammar.y:1132
 		{
 			yrVAL.exprs = []ast.Expression{yrDollar[1].ident}
 		}
 	case 115:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1061
+//line parser/grammar.y:1136
 		{
 			yrVAL.exprs = append(yrDollar[1].exprs, yrDollar[3].ident)
 		}
 	case 116:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1069
+//line parser/grammar.y:1144
 		{
 			yrVAL.ident = &ast.Identifier{Identifier: yrDollar[1].s}
 		}
 	case 117:
 		yrDollar = yrS[yrpt-2 : yrpt+1]
-//line parser/grammar.y:1073
+//line parser/grammar.y:1148
 		{
 			yrVAL.ident = &ast.Identifier{Identifier: yrDollar[1].s + "*"}
 		}
 	case 118:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1081
+//line parser/grammar.y:1156
 		{
 			yrVAL.expr = yrDollar[1].expr
 		}
 	case 119:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1085
+//line parser/grammar.y:1160
 		{
 			yrVAL.expr = ast.KeywordAll
 		}
 	case 120:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1089
+//line parser/grammar.y:1164
 		{
 			yrVAL.expr = ast.KeywordAny
 		}
 	case 121:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1093
+//line parser/grammar.y:1168
 		{
 			yrVAL.expr = ast.KeywordNone
 		}
 	case 122:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1101
+//line parser/grammar.y:1176
 		{
 			yrVAL.ss = []string{yrDollar[1].s}
 		}
 	case 123:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1105
+//line parser/grammar.y:1180
 		{
 			yrVAL.ss = append(yrDollar[1].ss, yrDollar[3].s)
 		}
 	case 124:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1112
+//line parser/grammar.y:1187
 		{
 			yrVAL.node = yrDollar[1].expr
 		}
 	case 125:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1116
+//line parser/grammar.y:1191
 		{
 			yrVAL.node = yrDollar[1].node
 		}
 	case 126:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1124
+//line parser/grammar.y:1199
 		{
 			yrVAL.expr = &ast.Group{yrDollar[2].expr}
 		}
 	case 127:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1128
+//line parser/grammar.y:1203
 		{
 			yrVAL.expr = ast.KeywordFilesize
 		}
 	case 128:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1132
+//line parser/grammar.y:1207
 		{
 			yrVAL.expr = ast.KeywordEntrypoint
 		}
 	case 129:
 		yrDollar = yrS[yrpt-4 : yrpt+1]
-//line parser/grammar.y:1136
+//line parser/grammar.y:1211
 		{
 			yrVAL.expr = &ast.FunctionCall{
 				Callable:  &ast.Identifier{Identifier: yrDollar[1].s},
@@ -1900,19 +1975,19 @@ yrdefault:
 		}
 	case 130:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1143
+//line parser/grammar.y:1218
 		{
 			yrVAL.expr = &ast.LiteralInteger{yrDollar[1].i64}
 		}
 	case 131:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1147
+//line parser/grammar.y:1222
 		{
 			yrVAL.expr = &ast.LiteralFloat{yrDollar[1].f64}
 		}
 	case 132:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1151
+//line parser/grammar.y:1226
 		{
 			if err := validateUTF8(yrDollar[1].s); err != nil {
 				return asLexer(yrlex).setError(
@@ -1923,136 +1998,190 @@ yrdefault:
 		}
 	case 133:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1160
+//line parser/grammar.y:1235
 		{
+			identifier := strings.TrimPrefix(yrDollar[1].s, "#")
+			if identifier != "" {
+				lexer := asLexer(yrlex)
+				if _, ok := lexer.strings[identifier]; !ok {
+					return lexer.setError(
+						gyperror.UndefinedStringIdentifierError,
+						`undefined string identifier: %s`, yrDollar[1].s)
+				}
+			}
 			yrVAL.expr = &ast.StringCount{
-				Identifier: strings.TrimPrefix(yrDollar[1].s, "#"),
+				Identifier: identifier,
 				In:         yrDollar[3].rng,
 			}
 		}
 	case 134:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1167
+//line parser/grammar.y:1251
 		{
+			identifier := strings.TrimPrefix(yrDollar[1].s, "#")
+			if identifier != "" {
+				lexer := asLexer(yrlex)
+				if _, ok := lexer.strings[identifier]; !ok {
+					return lexer.setError(
+						gyperror.UndefinedStringIdentifierError,
+						`undefined string identifier: %s`, yrDollar[1].s)
+				}
+			}
 			yrVAL.expr = &ast.StringCount{
-				Identifier: strings.TrimPrefix(yrDollar[1].s, "#"),
+				Identifier: identifier,
 			}
 		}
 	case 135:
 		yrDollar = yrS[yrpt-4 : yrpt+1]
-//line parser/grammar.y:1173
+//line parser/grammar.y:1266
 		{
+			identifier := strings.TrimPrefix(yrDollar[1].s, "@")
+			if identifier != "" {
+				lexer := asLexer(yrlex)
+				if _, ok := lexer.strings[identifier]; !ok {
+					return lexer.setError(
+						gyperror.UndefinedStringIdentifierError,
+						`undefined string identifier: %s`, yrDollar[1].s)
+				}
+			}
 			yrVAL.expr = &ast.StringOffset{
-				Identifier: strings.TrimPrefix(yrDollar[1].s, "@"),
+				Identifier: identifier,
 				Index:      yrDollar[3].expr,
 			}
 		}
 	case 136:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1180
+//line parser/grammar.y:1282
 		{
+			identifier := strings.TrimPrefix(yrDollar[1].s, "@")
+			if identifier != "" {
+				lexer := asLexer(yrlex)
+				if _, ok := lexer.strings[identifier]; !ok {
+					return lexer.setError(
+						gyperror.UndefinedStringIdentifierError,
+						`undefined string identifier: %s`, yrDollar[1].s)
+				}
+			}
 			yrVAL.expr = &ast.StringOffset{
-				Identifier: strings.TrimPrefix(yrDollar[1].s, "@"),
+				Identifier: identifier,
 			}
 		}
 	case 137:
 		yrDollar = yrS[yrpt-4 : yrpt+1]
-//line parser/grammar.y:1186
+//line parser/grammar.y:1297
 		{
+			identifier := strings.TrimPrefix(yrDollar[1].s, "!")
+			if identifier != "" {
+				lexer := asLexer(yrlex)
+				if _, ok := lexer.strings[identifier]; !ok {
+					return lexer.setError(
+						gyperror.UndefinedStringIdentifierError,
+						`undefined string identifier: %s`, yrDollar[1].s)
+				}
+			}
 			yrVAL.expr = &ast.StringLength{
-				Identifier: strings.TrimPrefix(yrDollar[1].s, "!"),
+				Identifier: identifier,
 				Index:      yrDollar[3].expr,
 			}
 		}
 	case 138:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1193
+//line parser/grammar.y:1313
 		{
+			identifier := strings.TrimPrefix(yrDollar[1].s, "!")
+			if identifier != "" {
+				lexer := asLexer(yrlex)
+				if _, ok := lexer.strings[identifier]; !ok {
+					return lexer.setError(
+						gyperror.UndefinedStringIdentifierError,
+						`undefined string identifier: %s`, yrDollar[1].s)
+				}
+			}
 			yrVAL.expr = &ast.StringLength{
 				Identifier: strings.TrimPrefix(yrDollar[1].s, "!"),
 			}
 		}
 	case 139:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1199
+//line parser/grammar.y:1328
 		{
 			yrVAL.expr = yrDollar[1].expr
 		}
 	case 140:
 		yrDollar = yrS[yrpt-2 : yrpt+1]
-//line parser/grammar.y:1203
+//line parser/grammar.y:1332
 		{
 			yrVAL.expr = &ast.Minus{yrDollar[2].expr}
 		}
 	case 141:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1207
+//line parser/grammar.y:1336
 		{
 			yrVAL.expr = operation(ast.OpAdd, yrDollar[1].expr, yrDollar[3].expr)
 		}
 	case 142:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1211
+//line parser/grammar.y:1340
 		{
 			yrVAL.expr = operation(ast.OpSub, yrDollar[1].expr, yrDollar[3].expr)
 		}
 	case 143:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1215
+//line parser/grammar.y:1344
 		{
 			yrVAL.expr = operation(ast.OpMul, yrDollar[1].expr, yrDollar[3].expr)
 		}
 	case 144:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1219
+//line parser/grammar.y:1348
 		{
 			yrVAL.expr = operation(ast.OpDiv, yrDollar[1].expr, yrDollar[3].expr)
 		}
 	case 145:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1223
+//line parser/grammar.y:1352
 		{
 			yrVAL.expr = operation(ast.OpMod, yrDollar[1].expr, yrDollar[3].expr)
 		}
 	case 146:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1227
+//line parser/grammar.y:1356
 		{
 			yrVAL.expr = operation(ast.OpBitXor, yrDollar[1].expr, yrDollar[3].expr)
 		}
 	case 147:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1231
+//line parser/grammar.y:1360
 		{
 			yrVAL.expr = operation(ast.OpBitAnd, yrDollar[1].expr, yrDollar[3].expr)
 		}
 	case 148:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1235
+//line parser/grammar.y:1364
 		{
 			yrVAL.expr = operation(ast.OpBitOr, yrDollar[1].expr, yrDollar[3].expr)
 		}
 	case 149:
 		yrDollar = yrS[yrpt-2 : yrpt+1]
-//line parser/grammar.y:1239
+//line parser/grammar.y:1368
 		{
 			yrVAL.expr = &ast.BitwiseNot{yrDollar[2].expr}
 		}
 	case 150:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1243
+//line parser/grammar.y:1372
 		{
 			yrVAL.expr = operation(ast.OpShiftLeft, yrDollar[1].expr, yrDollar[3].expr)
 		}
 	case 151:
 		yrDollar = yrS[yrpt-3 : yrpt+1]
-//line parser/grammar.y:1247
+//line parser/grammar.y:1376
 		{
 			yrVAL.expr = operation(ast.OpShiftRight, yrDollar[1].expr, yrDollar[3].expr)
 		}
 	case 152:
 		yrDollar = yrS[yrpt-1 : yrpt+1]
-//line parser/grammar.y:1251
+//line parser/grammar.y:1380
 		{
 			yrVAL.expr = yrDollar[1].reg
 		}
diff --git a/tests/grammar_test.go b/tests/grammar_test.go
index 4e2dd47..2d8bd8b 100644
--- a/tests/grammar_test.go
+++ b/tests/grammar_test.go
@@ -286,6 +286,7 @@ rule AND_OR_PRECEDENCE_NO_PARENS {
     $foo1 = "foo1"
     $foo2 = /foo2/
     $foo3 = { AA BB CC }
+    $foo4 = "I AM A STRING! ;)"
   condition:
     $foo1 or $foo2 or $foo3 and $foo4
 }
@@ -295,6 +296,7 @@ rule AND_OR_PRECEDENCE_PARENS {
     $foo1 = "foo1"
     $foo2 = /foo2/
     $foo3 = { AA BB CC }
+    $foo4 = "I AM A STRING! ;)"
   condition:
     ($foo1 or $foo2 or $foo3) and $foo4
 }
@@ -459,6 +461,237 @@ func TestDuplicateStringModifiers(t *testing.T) {
 	}
 }
 
+func TestUndefinedStringIdentifier(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_IDENTIFIER {
+		strings:
+			$a = "AXSERS"
+		condition:
+			$s
+	}`)
+	if assert.Error(t, err) {
+		assert.Equal(t, `line 7: undefined string identifier: $s`, err.Error())
+	}
+}
+
+func TestUndefinedStringKeywordNoStrings(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_KEYWORD_NO_STRINGS {
+		condition:
+			1 of them
+	}`)
+	if assert.Error(t, err) {
+		assert.Equal(t, `line 4: undefined string identifier: them`, err.Error())
+	}
+}
+
+func TestUndefinedStringIdentifierWildcard(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_IDENTIFIER_WITH_WILDCARD {
+		condition:
+			1 of ($s*)
+	}`)
+	if assert.Error(t, err) {
+		assert.Equal(t, `line 4: undefined string identifier: $s*`, err.Error())
+	}
+}
+
+// Make sure wildcard expansion still works.
+func TestUndefinedStringIdentifierWildcardExpansion(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_IDENTIFIER_WITH_WILDCARD_2 {
+		strings:
+			$s0 = "AXSERS"
+		condition:
+			1 of ($s*) and $x
+	}`)
+	if assert.Error(t, err) {
+		assert.Equal(t, `line 7: undefined string identifier: $x`, err.Error())
+	}
+}
+
+func TestUndefinedStringCount(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_COUNT {
+		condition:
+			#s > 0
+	}`)
+	if assert.Error(t, err) {
+		assert.Equal(t, `line 4: undefined string identifier: #s`, err.Error())
+	}
+}
+
+// Make sure anonymous string counts are allowed.
+func TestUndefinedStringCountAnonymous(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_COUNT_ANONYMOUS {
+		strings:
+			$s = "AXSERS"
+		condition:
+			for any of them: (# > 10)
+	}`)
+	assert.NoError(t, err)
+}
+
+func TestUndefinedStringOffset(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_OFFSET {
+		condition:
+			@s[0] > 0
+	}`)
+	if assert.Error(t, err) {
+		assert.Equal(t, `line 4: undefined string identifier: @s`, err.Error())
+	}
+}
+
+// Make sure anonymous string offsets are allowed.
+func TestUndefinedStringOffsetAnonymous(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_OFFSET_ANONYMOUS {
+		strings:
+			$s = "AXSERS"
+		condition:
+			for any of them: (@ > 10)
+	}`)
+	assert.NoError(t, err)
+}
+
+func TestUndefinedStringOffsetAnonymousExpression(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_OFFSET_ANONYMOUS_EXPRESSION {
+		strings:
+			$s = "AXSERS"
+		condition:
+			for any of them: (@[1] > 10)
+	}`)
+	assert.NoError(t, err)
+}
+
+func TestUndefinedStringLength(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_LENGTH {
+		condition:
+			!s == 40
+	}`)
+	if assert.Error(t, err) {
+		assert.Equal(t, `line 4: undefined string identifier: !s`, err.Error())
+	}
+}
+
+// Make sure anonymous string lengths are allowed.
+func TestUndefinedStringLengthAnonymous(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_LENGTH_ANONYMOUS {
+		strings:
+			$s = "AXSERS"
+		condition:
+			for any of them: (! > 5)
+	}`)
+	assert.NoError(t, err)
+}
+
+func TestUndefinedStringLengthAnonymousExpression(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_LENGTH_ANONYMOUS_EXPRESSION {
+		strings:
+			$s = "AXSERS"
+		condition:
+			for any of them: (![1] > 5)
+	}`)
+	assert.NoError(t, err)
+}
+
+func TestUndefinedStringAt(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_AT {
+		condition:
+			$s at 10
+	}`)
+	if assert.Error(t, err) {
+		assert.Equal(t, `line 5: undefined string identifier: $s`, err.Error())
+	}
+}
+
+// Make sure anonymous string at are allowed.
+func TestUndefinedStringAtAnonymous(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_AT_ANONYMOUS {
+		strings:
+			$s = "AXSERS"
+		condition:
+			for any of them: ($ at 5)
+	}`)
+	assert.NoError(t, err)
+}
+
+func TestUndefinedStringInRange(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_IN_RANGE {
+		condition:
+			$s in (0..10)
+	}`)
+	if assert.Error(t, err) {
+		assert.Equal(t, `line 4: undefined string identifier: $s`, err.Error())
+	}
+}
+
+// Make sure anonymous string in range is allowed.
+func TestUndefinedStringInRangeAnonymous(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_IN_RANGE_ANONYMOUS {
+		strings:
+		  $s = "AXSERS"
+		condition:
+		    for any of them: ($ in (0..10))
+	}`)
+	assert.NoError(t, err)
+}
+
+func TestUndefinedStringInRangeAnonymousExpression(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_IN_RANGE_ANONYMOUS_EXPRESSION {
+		strings:
+			$s = "AXSERS"
+		condition:
+			for any of them: ($ in (0..10))
+	}`)
+	assert.NoError(t, err)
+}
+
+func TestUndefinedStringCountInRange(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_COUNT_IN_RANGE {
+		condition:
+			#s in (0..10) == 2
+	}`)
+	if assert.Error(t, err) {
+		assert.Equal(t, `line 4: undefined string identifier: #s`, err.Error())
+	}
+}
+
+// Make sure anonymous string count in range is allowed.
+func TestUndefinedStringCountInRangeAnonymous(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_COUNT_IN_RANGE_ANONYMOUS {
+		strings:
+			$s = "AXSERS"
+		condition:
+			for any of them: (# in (0..10) == 2)
+	}`)
+	assert.NoError(t, err)
+}
+
+func TestUndefinedStringEnumerationAnonymous(t *testing.T) {
+	_, err := gyp.ParseString(`
+	rule UNDEFINED_STRING_ENUMERATION_ANONYMOUS {
+		condition:
+			any of ($)
+	}`)
+	if assert.Error(t, err) {
+		assert.Equal(t, `line 4: undefined string identifier: $`, err.Error())
+	}
+}
+
 func TestBase64AlphabetLength(t *testing.T) {
 	_, err := gyp.ParseString(`
 	rule BASE64 {