Discord Bot with Ollama API

This project is a Discord bot that integrates with an Express server to generate AI-powered responses using Ollama. Users can send messages with a prefix (!) in a Discord server, and the bot will generate a response using the DeepSeek model.

Features

• AI-generated responses using Ollama API
• Express backend for handling AI requests
• Discord bot using discord.js
• Configurable command prefix

Prerequisites

Ensure you have the following installed:

• Node.js
• Discord Developer Account
• Ollama installed in your system.

Set-up

The installation can be divided into two steps, since the project is currently not deployed, users will have to run to run their own backend to run this chatbot.

First Step (Setting up a bot on discord):

---

1. Create a New Bot

1. Go to the Discord Developer Portal.
2. Click New Application → Enter a name → Click Create.

---

2. Configure Bot Settings

Enable Privileged Intents

1. Scroll down to Privileged Gateway Intents.

2. Enable MESSAGE CONTENT INTENT.

3. Click Save Changes.

   

---

3. Copy the Bot Token

1. Under Bot, scroll down to the Token section.

2. Click Reset Token if needed.

3. Copy and store it in the .env file (explained later).

   

---

4. Add the Bot to a Discord Server

1. Navigate to OAuth2 → URL Generator.
2. Select bot and applications.commands.

3. Under Bot Permissions, select:

   • Send Messages
   • Read Message History
   • Connect
   • Use Slash Commands (optional)

4. Copy the generated URL and open it in your server.

5. Select your server and authorize the bot.

---

Second Step (Running the backend of the bot so that it gives responses on the channel):

Start Ollama

Before running the bot, make sure Ollama is installed and running (if you do not have Ollama installed in your system, check out the Prerequisites section):

ollama serve

1. Clone the repository

git clone https://github.com/yourusername/discord-ollama-bot.git
cd discord-ollama-bot

2. Install dependencies

npm install

3. Set up environment variables

Create a .env file in the root directory and add the following:

DISCORD_TOKEN=your_discord_bot_token
PREFIX=! (this prefix can be the user's choice)

Replace your_discord_bot_token with your actual bot token from the Discord Developer Portal.

4. Start the Express server

Run the following command to start the backend:

node index.js

5. Start the Discord bot

Run:

npm run bot

Usage

• Type !your-message in a Discord server where the bot is added.
• The bot will send an AI-generated response based on the message.

Future Updates

Adding vector db to increase llm memory

[Bot] <-> [Express App] <-> [Vector Search] <-> [Qdrant]
                                   |
                                [Ollama]

Deployment

• Deploy the Express backend to services like Vercel, Railway, or a VPS.
• Host the Discord bot on Replit, Railway, or a VPS for 24/7 uptime.

Future Improvements

• Add slash commands support.
• Improve response handling and error messages.
• Deploy backend to a cloud-based solution.

📝 Available Commands

• /ping - Check bot latency.

License

This project is open-source under the MIT License.

---

Need Help?

Feel free to open an issue or reach out!