NeuroCord

A production-ready Discord AI chatbot powered by GROQ (llama-3.1-8b-instant), built with Express.js and the Discord Interactions API (slash commands over HTTP — no WebSocket required).

/ask:Explain recursion simply

---

How It Works

1. User types /ask:<anything> in Discord
2. Discord sends an HTTP POST to your server's /interactions endpoint
3. The server immediately responds with a deferred response (type 5) → Discord shows "Bot is thinking…"
4. The server asynchronously calls GROQ API to get the AI response
5. The server PATCHes the original deferred message with the AI reply

This architecture means the server only needs to be alive when Discord is actively sending a request — perfect for Render free tier.

---

Project Structure

neurocord/
├── src/
│   ├── index.js          # Express server & interaction router
│   ├── verify.js         # Discord request signature verification
│   ├── groq.js           # GROQ Chat Completions client
│   ├── discord.js        # Follow-up message helpers
│   └── commands/
│       └── ask.js        # /ask command handler
├── scripts/
│   └── register.js       # One-time slash command registration
├── .env.example          # Environment variable template
├── render.yaml           # Render deployment config
└── package.json

---

Setup Guide

1. Create a Discord Application

1. Go to Discord Developer Portal
2. Click New Application → give it a name (e.g., "NeuroCord")
3. Navigate to General Information and copy your Application ID and Public Key
4. Navigate to Bot → click Add Bot → copy the Bot Token
5. Under Bot → Privileged Gateway Intents, enable what you need (none required for slash commands)

2. Get a GROQ API Key

1. Go to console.groq.com
2. Sign up / log in → API Keys → Create API Key
3. Copy the key

3. Configure Environment Variables

cp .env.example .env

Fill in .env:

DISCORD_PUBLIC_KEY=<from Discord Developer Portal → General Information>
DISCORD_BOT_TOKEN=<from Discord Developer Portal → Bot>
DISCORD_APPLICATION_ID=<from Discord Developer Portal → General Information>
GROQ_API_KEY=<from console.groq.com>
PORT=3000

4. Install Dependencies

npm install

5. Register the /ask Slash Command

Run this once to register the slash command with Discord:

npm run register

Output:

✅ Registered 1 command(s):
   /ask — Ask the AI anything!

Note: Global commands can take up to 1 hour to propagate. For instant testing, add a DISCORD_GUILD_ID to .env and modify scripts/register.js to use the guild-scoped endpoint.

6. Run Locally (with ngrok)

npm start
# In another terminal:
npx ngrok http 3000

Copy the ngrok HTTPS URL (e.g., https://abc123.ngrok.io).

7. Set the Interactions Endpoint URL

1. In Discord Developer Portal → your app → General Information
2. Paste your URL + /interactions into Interactions Endpoint URL
   • e.g., https://abc123.ngrok.io/interactions
3. Click Save Changes — Discord will send a PING. If the endpoint is valid, it saves successfully.

---

Deploy to Render

1. Push this repo to GitHub
2. Go to render.com → New Web Service → connect your repo
3. Render auto-detects render.yaml and configures the service
4. In Environment, add your secret values:
   • DISCORD_PUBLIC_KEY
   • DISCORD_BOT_TOKEN
   • DISCORD_APPLICATION_ID
   • GROQ_API_KEY
5. Deploy — copy your Render URL (e.g., https://neurocord.onrender.com)
6. Set Interactions Endpoint URL to https://neurocord.onrender.com/interactions

Render free tier note: The service will sleep after 15 minutes of inactivity. It wakes automatically when Discord sends the next interaction — no keep-alive needed since Discord retries.

---

Invite the Bot to Your Server

In Discord Developer Portal → OAuth2 → URL Generator:

• Scopes: bot, applications.commands
• Bot Permissions: Send Messages

Copy the generated URL, open it in a browser, and add the bot to your server.

---

Usage

/ask:What is the meaning of life?
/ask:Write a Python function to reverse a string
/ask:Explain quantum entanglement in simple terms