fix(cmd): sanitize followup reminder UTF-8 truncation (#609)

Prevent invalid UTF-8 from being persisted when auto-setting followups by sanitizing message content and truncating by rune count. Add regression tests for emoji truncation and malformed byte sequences.

cmd/gateway_consumer.go

@@ -8,6 +8,7 @@ import (
 	"strings"
 	"sync"
 	"time"
+	"unicode/utf8"
 
 	"github.com/google/uuid"
 
@@ -212,8 +213,14 @@ func truncateForReminder(content string, maxLen int) string {
 	// Use last non-empty line as it's typically the most relevant.
 	lines := strings.Split(strings.TrimSpace(content), "\n")
 	msg := lines[len(lines)-1]
-	if len(msg) > maxLen {
-		msg = msg[:maxLen] + "..."
+	// Ensure we only persist valid UTF-8 into PostgreSQL.
+	msg = strings.ToValidUTF8(msg, "")
+	if maxLen <= 0 {
+		return msg
+	}
+	if utf8.RuneCountInString(msg) > maxLen {
+		r := []rune(msg)
+		msg = string(r[:maxLen]) + "..."
 	}
 	return msg
 }

cmd/gateway_consumer_followup_test.go

@@ -0,0 +1,34 @@
+package cmd
+
+import (
+	"strings"
+	"testing"
+	"unicode/utf8"
+)
+
+func TestTruncateForReminder_TruncatesByRuneAndKeepsUTF8Valid(t *testing.T) {
+	input := strings.Repeat("a", 199) + "✌️"
+
+	got := truncateForReminder(input, 200)
+	if !utf8.ValidString(got) {
+		t.Fatalf("truncateForReminder() produced invalid UTF-8: %q", got)
+	}
+	if !strings.HasSuffix(got, "...") {
+		t.Fatalf("truncateForReminder() = %q, want suffix ...", got)
+	}
+	if strings.Contains(got, "\uFFFD") {
+		t.Fatalf("truncateForReminder() introduced replacement rune: %q", got)
+	}
+}
+
+func TestTruncateForReminder_StripsInvalidUTF8(t *testing.T) {
+	invalid := string([]byte{'a', 0xef, 0xb8, '.', 'b'})
+
+	got := truncateForReminder(invalid, 200)
+	if !utf8.ValidString(got) {
+		t.Fatalf("truncateForReminder() produced invalid UTF-8: %q", got)
+	}
+	if strings.ContainsRune(got, '\uFFFD') {
+		t.Fatalf("truncateForReminder() = %q, want no replacement rune", got)
+	}
+}