Tater is a Discord bot that integrates with Ollama to provide a variety of AI-powered tools, and now it also comes with a web UI for internal private chat. Whether you're on Discord or using the WebUI, Tater is at your service.
- Conversation Continuity: Maintains context using Redis.
- Ollama Integration: Utilizes Ollama for AI responses.
- Chat Responses: Generates AI responses, waiting messages, and friendly error messages.
- Note:
- You DO NOT need to use a model that is tagged with tools to use this, test different models if the one you are using isnt trigging the plugins, but it does not need to be under Ollama model with tools support list.
- Tater currently recommends using gemma3:27b
Below are the tools available to you. Simply ask Tater to perform these tasks—no slash commands or specific key terms are required:
-
YouTube Video Summaries:
Extracts YouTube video IDs, fetches summaries, and sends formatted responses. -
Web Summaries:
Summarizes webpages or articles. -
Image Generation:
Generates images based on text prompts using Automatic111/ComfyUI. -
Premiumize.me Integration:
- Checks if a given URL is cached on Premiumize.me and retrieves download links.
- Processes torrent files to extract the torrent hash, checks cache status, and retrieves download links.
-
RSS Feed Monitoring:
Automatically monitors RSS feeds for new articles and announces summaries to RESPONSE_CHANNEL when new articles are published. This integration includes three tools:- Watch Feed: Add an RSS feed to be monitored.
- Unwatch Feed: Remove an RSS feed from monitoring.
- List Feeds: List all currently watched RSS feeds.
-
Web Search:
Searches the web for additional or up-to-date information when needed. If the AI determines that it lacks sufficient knowledge or context to answer a query, it can trigger a web search to retrieve current information and use it to generate a final, accurate answer.
For a detailed example on how to create your own plugin, check out the plugin template.
The web UI provides a Streamlit-based interface to interact with Tater. Key features include:
-
Interactive Chat:
Engage with Tater via an intuitive chat interface that supports chat history, file attachments, and tool function calls. -
Discord Bot Settings:
Configure Discord-related settings directly from the UI. In the sidebar, under the "Discord Settings" expander, you can set:- DISCORD_TOKEN: Your bot's token.
- ADMIN_USER_ID: The Discord ID of the admin.
- RESPONSE_CHANNEL_ID: The channel ID where Tater sends responses.
- RSS_CHANNEL_ID: The channel ID for RSS feed announcements.
-
Chat Settings:
Customize your user avatar and username. These settings allow you to personalize your experience and are stored persistently in Redis. -
File Attachments:
Upload files or images directly from the sidebar, which are then processed accordingly by Tater (for example, torrent files for Premiumize functions).
This unified interface lets you manage both chat interactions and Discord settings in one place, making it easy to deploy and maintain Tater across both platforms.
- RSS Feed Management (Discord-only):
- Watch Feeds: Add an RSS feed to the watch list.
- Unwatch Feeds: Remove an RSS feed.
- List Feeds: List all currently watched RSS feeds.
- (RSS feed announcements post to a dedicated Discord channel.)
- Python 3.11
- Docker (optional, for containerized deployment)
- Clone the Repository
git clone https://github.com/MasterPhooey/Tater-Discord-WebUI.git- Navigate to the Project Directory
cd Tater-Discord-WebUI- Install Dependencies
Using pip, run:
pip install -r requirements.txt- Configure Environment Variables
Create a .env file in the root directory with the following variables:
OLLAMA_HOST=127.0.0.1
OLLAMA_PORT=11434
OLLAMA_MODEL=gemma3:27b
CONTEXT_LENGTH=10000
REDIS_HOST=127.0.0.1
REDIS_PORT=6379- Run the Web UI
Launch the web UI using Streamlit:
streamlit run webui.pyFor users who prefer not to build the Docker image locally, a prebuilt image is available on Docker Hub.
Pull the prebuilt image with the following command:
docker pull masterphooey/tater-webuiRun the container with the following command. This example maps port 8501 on your host to port 8501 in the container (where the Streamlit web UI runs):
docker run -d --name tater_webui -p 8501:8501 masterphooey/tater-webuiEnsure you supply the required environment variables. You can pass these using the -e flag when starting the container. For example:
docker run -d --name tater_webui \
-p 8501:8501 \
-e OLLAMA_HOST=127.0.0.1 \
-e OLLAMA_PORT=11434 \
-e OLLAMA_MODEL=gemma3:27b \
-e CONTEXT_LENGTH=10000 \
-e REDIS_HOST=redis \
-e REDIS_PORT=6379 \
masterphooey/tater-webuiOnce the container is running, open your browser and navigate to:
The Streamlit-based web UI will be available for interacting with Tater.
If you're running Redis directly, you can ensure data persistence by configuring Redis's built-in persistence mechanisms. Redis supports two primary methods:
Redis periodically saves snapshots of your dataset to a file (typically named dump.rdb). You can control this behavior in your redis.conf file using save directives. For example:
# Save a snapshot every 900 seconds if at least 1 key changed.
save 900 1
# Save a snapshot every 300 seconds if at least 10 keys changed.
save 300 10
# Save a snapshot every 60 seconds if at least 10000 keys changed.
save 60 10000Redis can log every write operation to an AOF file, which can be replayed to reconstruct the dataset. To enable AOF, add the following lines to your redis.conf:
appendonly yes
appendfilename "appendonly.aof"Alternatively, you can launch Redis with these persistence settings directly from the terminal (without modifying your redis.conf):
redis-server --save "900 1" --save "300 10" --save "60 10000" --appendonly yes --appendfilename "appendonly.aof"