diff --git a/donut.c b/donut.c
index d93e7001..2b8ac990 100644
--- a/donut.c
+++ b/donut.c
@@ -2085,6 +2085,9 @@ static int validate_format(opt_arg *arg, void *args) {
       } else
       if(!strcasecmp("hex", str)) {
         arg->u32 = DONUT_FORMAT_HEX;
+      } else
+      if(!strcasecmp("uuid", str)) {
+        arg->u32 = DONUT_FORMAT_UUID;
       }
     }
     // validate
@@ -2097,6 +2100,7 @@ static int validate_format(opt_arg *arg, void *args) {
       case DONUT_FORMAT_POWERSHELL:
       case DONUT_FORMAT_CSHARP:
       case DONUT_FORMAT_HEX:
+      case DONUT_FORMAT_UUID:
         break;
       default: {
         printf("WARNING: Invalid format specified: %"PRId32" -- setting to binary.\n", arg->u32);
@@ -2150,7 +2154,7 @@ static void usage (void) {
     printf("                   -PIC/SHELLCODE OPTIONS-\n\n");    
     printf("       -a,--arch: <arch>,--cpu: <arch>         Target architecture : 1=x86, 2=amd64, 3=x86+amd64(default).\n");
     printf("       -o,--output: <path>                     Output file to save loader. Default is \"loader.bin\"\n");
-    printf("       -f,--format: <format>                   Output format. 1=Binary (default), 2=Base64, 3=C, 4=Ruby, 5=Python, 6=Powershell, 7=C#, 8=Hex\n");
+    printf("       -f,--format: <format>                   Output format. 1=Binary (default), 2=Base64, 3=C, 4=Ruby, 5=Python, 6=Powershell, 7=C#, 8=Hex, 9=UUID\n");
     printf("       -y,--fork: <offset>                     Create a new thread for the loader and continue execution at <offset> relative to the host process's executable.\n");
     printf("       -x,--exit: <action>                     Exit behaviour. 1=Exit thread (default), 2=Exit process, 3=Do not exit or cleanup and block indefinitely\n\n");
     
diff --git a/format.c b/format.c
index 9cb0882f..b2a315da 100644
--- a/format.c
+++ b/format.c
@@ -243,25 +243,27 @@ int hex_template(void * pic, uint32_t pic_len, FILE* fd){
     return DONUT_ERROR_OK;
 }
 
-static int uuid_null[16] = { 0 };
-
 int uuid_template(void * pic, uint32_t pic_len, FILE* fd){
     uint32_t rem;
     uint32_t j;
     uint32_t base;
     uint8_t *p = (uint8_t*)pic;
     uint32_t len = pic_len;
+    uint32_t chunks = (len + 15) / 16;
+    uint8_t last_chunk[16];
 
-    //Ensure there are enough bytes
-    rem = len % 16;
-    if(rem != 0){
-        pic = realloc(pic, len+rem);
-        memcpy(p + len, uuid_null, rem);
-        len+=rem;
-    }
-
-    for(j=0; j < len/16; j++){
+    for(j=0; j < chunks; j++){
         base = j*16;
+
+        // last chunk have less than 16bytes
+        if (base+16 >= pic_len) {
+            rem = len % 16;
+            memcpy(last_chunk, p+base, rem);
+            memset(last_chunk+rem, 0xcc, 16-rem);
+            p = last_chunk;
+            base = 0;
+        }
+
         fprintf(fd, "%02x%02x%02x%02x-", p[base+3], p[base+2], p[base+1], p[base]);
         fprintf(fd, "%02x%02x-", p[base+5], p[base+4]);
         fprintf(fd, "%02x%02x-", p[base+7], p[base+6]);