diff --git a/src/bot.py b/src/bot.py
index bfa1970..8565d92 100644
--- a/src/bot.py
+++ b/src/bot.py
@@ -1,12 +1,10 @@
-import discord
 from discord.ext import commands
 import os
 
+bot = commands.Bot(command_prefix='!')
 
-bot = commands.Bot(command_prefix='¡')
-
-TOKEN = open('token.txt', 'r').read()
-
+with open('token.txt', 'r') as token_file:
+    TOKEN = token_file.read()
 
 for file in os.listdir('./funcionalidades'):
     if file.endswith('.py'):
diff --git a/src/funcionalidades/ban_hammer.py b/src/funcionalidades/ban_hammer.py
index 0bcdcb9..9fa937b 100644
--- a/src/funcionalidades/ban_hammer.py
+++ b/src/funcionalidades/ban_hammer.py
@@ -1,6 +1,18 @@
-import discord
 from discord.ext import commands
 
+PERMITTED_ROLES_NAMES = ('Junta', 'Admin')
+
+
+def setup(bot):
+    bot.add_cog(BanHammer(bot))
+
+
+def have_permitted_rol(autor_roles):
+    for autor_rol in autor_roles:
+        if autor_rol.name in PERMITTED_ROLES_NAMES:
+            return True
+    return False
+
 
 class BanHammer(commands.Cog):
 
@@ -20,14 +32,27 @@ async def on_message(self, message):
         if message.author == self.bot.user:
             return
 
+        message_content = message.content
         forbidden_words_used = [
-            i for i in self.blacklist if message.content.casefold().count(i) > 0]
+            i for i in self.blacklist if message_content.casefold().count(i) > 0]
 
-        if(len(forbidden_words_used) > 0):
-            await message.delete()  # Delete the message
-            # Send a message on the same channel
-            await message.channel.send(message.author.mention + ", debes cuidar tu vocabulario, jovencito")
-            await message.author.send("El mensaje \n" + str(""f"```css\n{message.content}```""") + "no se ajusta a las normas, en los próximos mensajes no uses: " + str(forbidden_words_used).strip('[]'))  # Send a private message to the user
+        if len(forbidden_words_used) > 0:
+            censor_command = self.bot.get_command('censor')
+            censor_command_names = censor_command.aliases
+            censor_command_names.append(censor_command.name)
+            str_ = message_content.split()[0]
+            # If one with permitted roles is baning a word, we dont ban his message
+            if have_permitted_rol(message.author.roles) and str_[0] == self.bot.command_prefix \
+                    and str_[1:] in censor_command_names:
+                pass
+            else:
+                await message.delete()  # Delete the message
+                # Send a message on the same channel
+                await message.channel.send(message.author.mention + ", debes cuidar tu vocabulario, jovencito")
+                # Send a private message to the user
+                await message.author.send("El mensaje \n" + str(f'```diff\n-"{message_content}"```') +
+                                          "no se ajusta a las normas, en los próximos mensajes no uses: " +
+                                          str(forbidden_words_used).strip('[]'))
 
     @commands.command(name='censor')
     @commands.has_role('Junta')
@@ -36,7 +61,7 @@ async def ban_word(self, ctx):
         if word in self.blacklist:
             await ctx.send("La palabra ya estaba baneada")
         else:
-            with open("blacklist_insultos.txt", "a+") as file:
+            with open("blacklist_insultos.txt", "a") as file:
                 file.write("\n" + word)
             self.blacklist.append(word)
             await ctx.send('Palabra censurada correctamente :)')
@@ -59,7 +84,3 @@ async def unban_word(self, ctx):
     async def on_command_error(self, ctx, error):
         if isinstance(error, commands.errors.CheckFailure):
             await ctx.send('Lo siento, no es nada personal, pero no tienes permiso para hacer eso :)')
-
-
-def setup(bot):
-    bot.add_cog(BanHammer(bot))