diff --git a/corpus/env.txt b/corpus/env.txt
index 245599b..6079e83 100644
--- a/corpus/env.txt
+++ b/corpus/env.txt
@@ -26,6 +26,20 @@ ENV TEST_2=value\ 2
       name: (unquoted_string)
       value: (unquoted_string))))
 
+=========================================================
+Single-letter variable
+=========================================================
+
+ENV X=42
+
+---
+
+(source_file
+  (env_instruction
+    (env_pair
+      (unquoted_string)
+      (unquoted_string))))
+
 ================================================================================
 Multiple
 ================================================================================
diff --git a/grammar.js b/grammar.js
index f1a1a88..8acf84d 100644
--- a/grammar.js
+++ b/grammar.js
@@ -195,7 +195,7 @@ module.exports = grammar({
       ),
 
     _env_key: ($) =>
-      alias(/[a-zA-Z][a-zA-Z0-9_]*[a-zA-Z0-9]/, $.unquoted_string),
+      alias(/[a-zA-Z]([a-zA-Z0-9_]*[a-zA-Z0-9])?/, $.unquoted_string),
 
     expose_port: ($) => seq(/\d+/, optional(choice("/tcp", "/udp"))),
 
diff --git a/src/grammar.json b/src/grammar.json
index 948b931..c4d0d41 100644
--- a/src/grammar.json
+++ b/src/grammar.json
@@ -941,7 +941,7 @@
       "type": "ALIAS",
       "content": {
         "type": "PATTERN",
-        "value": "[a-zA-Z][a-zA-Z0-9_]*[a-zA-Z0-9]"
+        "value": "[a-zA-Z]([a-zA-Z0-9_]*[a-zA-Z0-9])?"
       },
       "named": true,
       "value": "unquoted_string"
diff --git a/src/parser.c b/src/parser.c
index 375ea04..734052d 100644
--- a/src/parser.c
+++ b/src/parser.c
@@ -1078,7 +1078,7 @@ static bool ts_lex(TSLexer *lexer, TSStateId state) {
           lookahead == '\r' ||
           lookahead == ' ') SKIP(5)
       if (('A' <= lookahead && lookahead <= 'Z') ||
-          ('a' <= lookahead && lookahead <= 'z')) ADVANCE(52);
+          ('a' <= lookahead && lookahead <= 'z')) ADVANCE(204);
       END_STATE();
     case 6:
       if (lookahead == '\n') ADVANCE(160);
@@ -1283,7 +1283,7 @@ static bool ts_lex(TSLexer *lexer, TSStateId state) {
           lookahead == '\r' ||
           lookahead == ' ') SKIP(22)
       if (('A' <= lookahead && lookahead <= 'Z') ||
-          ('a' <= lookahead && lookahead <= 'z')) ADVANCE(52);
+          ('a' <= lookahead && lookahead <= 'z')) ADVANCE(204);
       END_STATE();
     case 23:
       if (lookahead == '\n') ADVANCE(160);
@@ -1488,7 +1488,7 @@ static bool ts_lex(TSLexer *lexer, TSStateId state) {
           lookahead == '\r' ||
           lookahead == ' ') SKIP(45)
       if (('A' <= lookahead && lookahead <= 'Z') ||
-          ('a' <= lookahead && lookahead <= 'z')) ADVANCE(52);
+          ('a' <= lookahead && lookahead <= 'z')) ADVANCE(204);
       END_STATE();
     case 46:
       if (lookahead == '\\') ADVANCE(1);