diff --git a/src/cogs/chat.py b/src/cogs/chat.py
index 6b69064..76a5fd1 100644
--- a/src/cogs/chat.py
+++ b/src/cogs/chat.py
@@ -107,7 +107,7 @@ class Chat(discord.Cog):
             if message.content.startswith("botator!unban"):
                 user2ban = message.content.split(" ")[1]
                 await banusr.unbanuser(user2ban)
-                await message.channel.send(f"User {user2ban} unbanned !")
+                await message.chafnnel.send(f"User {user2ban} unbanned !")
                 return
         if str(message.author.id) in banusr.banend_users:
             await asyncio.sleep(2)
@@ -115,11 +115,6 @@ class Chat(discord.Cog):
             return
         await mp.chat_process(self, message)
 
-    @discord.slash_command(name="say", description="Say a message")
-    async def say(self, ctx: discord.ApplicationContext, message: str):
-        await ctx.respond("Message sent !", ephemeral=True)
-        await ctx.send(message)
-
     @discord.slash_command(name="redo", description="Redo a message")
     async def redo(self, ctx: discord.ApplicationContext):
         history = await ctx.channel.history(limit=2).flatten()
diff --git a/src/functionscalls.py b/src/functionscalls.py
index 2d94261..c7da912 100644
--- a/src/functionscalls.py
+++ b/src/functionscalls.py
@@ -5,6 +5,7 @@ import aiohttp
 import random
 import time
 
+from src.utils.misc import moderate
 from simpleeval import simple_eval
 from bs4 import BeautifulSoup
 from src.config import tenor_api_key
@@ -331,7 +332,7 @@ async def evaluate_math(
     return f"Result to math eval of {evaluable}: ```\n{str(result)}```"
 
 
-async def call_function(message: discord.Message, function_call):
+async def call_function(message: discord.Message, function_call, api_key):
     name = function_call.get("name", "")
     if name == "":
         raise FuntionCallError("No name provided")
@@ -341,6 +342,14 @@ async def call_function(message: discord.Message, function_call):
     if name not in functions_matching:
         raise FuntionCallError("Invalid function name")
     function = functions_matching[name]
+    if arguments.get("message", "") != "" and await moderate(
+        api_key=api_key, text=arguments["message"]
+    ):
+        return "Message blocked by the moderation system. Please try again."
+    if arguments.get("query", "") != "" and await moderate(
+        api_key=api_key, text=arguments["query"]
+    ):
+        return "Query blocked by the moderation system. If the user asked for something edgy, please tell them in a funny way that you won't do it, but do not specify that it was blocked by the moderation system."
     returnable = await function(message, arguments)
     return returnable
 
diff --git a/src/makeprompt.py b/src/makeprompt.py
index fb39257..f710269 100644
--- a/src/makeprompt.py
+++ b/src/makeprompt.py
@@ -6,7 +6,7 @@ import datetime
 import json
 
 from src.config import curs_data, max_uses, curs_premium, gpt_3_5_turbo_prompt
-from src.utils.misc import moderate
+from src.utils.misc import moderate, ModerationError, Hasher
 from src.utils.openaicaller import openai_caller
 from src.functionscalls import (
     call_function,
@@ -131,11 +131,12 @@ async def chatgpt_process(
         messages=msgs,
         functions=called_functions,
         function_call="auto",
+        user=Hasher(str(message.author.id)),  # for user banning in case of abuse
     )
     response = response["choices"][0]["message"]  # type: ignore
     if response.get("function_call"):
         function_call = response.get("function_call")
-        returned = await call_function(message, function_call)
+        returned = await call_function(message, function_call, api_key)
         if returned != None:
             msgs.append(
                 {
@@ -153,13 +154,24 @@ async def chatgpt_process(
             await chatgpt_process(self, msgs, message, api_key, prompt, model, depth)
     else:
         content = response.get("content", "")
-        while len(content) != 0:
-            if len(content) > 2000:
-                await message.channel.send(content[:2000])
-                content = content[2000:]
-            else:
-                await message.channel.send(content)
-                content = ""
+        if await moderate(api_key, content, error_call):
+            depth += 1
+            if depth > 2:
+                await message.channel.send(
+                    "Oh uh, it seems like i am answering recursively. I will stop now."
+                )
+                raise ModerationError("Too many recursive messages")
+            await chatgpt_process(
+                self, msgs, message, api_key, prompt, model, error_call, depth
+            )
+        else:
+            while len(content) != 0:
+                if len(content) > 2000:
+                    await message.channel.send(content[:2000])
+                    content = content[2000:]
+                else:
+                    await message.channel.send(content)
+                    content = ""
 
 
 async def chat_process(self, message):
diff --git a/src/utils/misc.py b/src/utils/misc.py
index 13f1aa9..15344e3 100644
--- a/src/utils/misc.py
+++ b/src/utils/misc.py
@@ -1,3 +1,5 @@
+import hashlib
+
 from src.utils.openaicaller import openai_caller
 
 
@@ -9,3 +11,20 @@ async def moderate(api_key, text, recall_func=None):
         input=text,
     )
     return response["results"][0]["flagged"]  # type: ignore
+
+
+class ModerationError(Exception):
+    pass
+
+
+class hasher:
+    def __init__(self):
+        self.hashes = {}
+
+    def __call__(self, text: str) -> str:
+        if self.hashes.get(text, None) is None:
+            self.hashes[text] = hashlib.sha256(text.encode()).hexdigest()
+        return self.hashes[text]
+
+
+Hasher = hasher()