Merge pull request #9 from rodonguyen/dev-discord-bot

Prototype a POC Discordbot using Interactions.py

.env.example

@@ -0,0 +1,4 @@
+DISCORD_BOT_TOKEN="EXAMPLETokenTTTTT.Grz0fK.IUi1IONPcLHZsnyjROTt8lR2fuYCeK4grzRoSQ"
+
+# THIS SHOULD BE ASSIGNED TO THE BOT'S `scopes` WHEN DEPLOYED
+DISCORD_GUILD_ID="99999999999999999999999"

.gitignore

@@ -158,3 +158,6 @@ cython_debug/
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
 #.idea/
+
+bin
+local_sandbox

discord_bot/bot.py

@@ -0,0 +1,66 @@
+import os
+from dotenv import load_dotenv
+import asyncio
+import interactions
+
+bot = interactions.Client(intents=interactions.Intents.DEFAULT)
+
+
+@interactions.listen()
+async def on_ready():
+    print("Ready")
+    print(f"This bot is owned by {bot.owner}")
+
+
+@interactions.listen()
+async def on_message_create(event):
+    print(f"message received: {event.message.content}")
+
+
+@interactions.slash_command(name="my_short_command", description="My first command :)")
+async def my_short_command(ctx: interactions.SlashContext):
+    await ctx.send("Hello World")
+
+
+@interactions.slash_command(name="my_long_command", description="My second command :)")
+async def my_long_command_function(ctx: interactions.SlashContext):
+    await ctx.defer()  # use defer for time-comsuming commands
+    await asyncio.sleep(600)
+    await ctx.send("Hello World")
+
+
+@interactions.slash_command(name="ask", description="Ask an LLM")
+@interactions.slash_option(
+    name="model",
+    description="Choose an LLM model",
+    required=True,
+    opt_type=interactions.OptionType.STRING,
+    autocomplete=True,
+)
+async def ask_model(ctx: interactions.SlashContext, model: str):
+    await ctx.send(f"You asked model {model}")
+
+
+@ask_model.autocomplete("model")
+async def autocomplete(ctx: interactions.AutocompleteContext):
+    string_option_input = ctx.input_text  # note: can be empty
+    print(f"input: {string_option_input}")
+    # you can use ctx.kwargs.get("name") to get the current state of other options - note they can be empty too
+    # make sure you respond within three seconds
+
+    choices = ["gpt3", "gpt4"]
+    filtered_choices = [choice for choice in choices if string_option_input in choice]
+
+    await ctx.send(
+        choices=[
+            {
+                "name": choice,
+                "value": choice,
+            }
+            for choice in filtered_choices
+        ]
+    )
+
+
+load_dotenv()
+bot.start(os.getenv("DISCORD_BOT_TOKEN"))

pyproject.toml

@@ -10,6 +10,7 @@ python = "^3.10"
 litellm = {extras = ["proxy"], version = "^1.23.7"}
 openai = "^1.11.1"
 python-dotenv = "^1.0.1"
+discord-py-interactions = "5.11.0"
 
 
 [build-system]