All apps · 0 apps
panko-gpt
Docker app from dracon's Repository
Overview
Readme
View on GitHubPankoGPT - AI Companion Platform

Currently in alpha
Create your own unique AI Companions on Discord, Telegram, WhatsApp and soon other platforms.
🚀 Overview
Create and Deploy AI Companions for your friends, and/or family across various messaging platforms, starting with WhatsApp, Discord and Telegram. With PankoGPT, you can:
- Easily Create Custom AI Companions: Deploy companions with specific goals and behaviors tailored to each need.
- Customizable Settings: Configure companion behavior by filling out straightforward forms—no coding required.
- Contextual Understanding: Define the scope and context in which your companion operates.
- Function Tools: Equip your companions with configurable tools like URL access, time fetching, and more.
Homepage

Create new companion platform selection

Create new discord companion form

Knowledgebase section

Selecting Function/Tools

Settings

Features
- User-Friendly Interface: Deploy custom companions for WhatsApp, Discord and Telegram (and soon other platforms) without the need for deep technical knowledge of their integration.
- Customizable Behavior: Fine-tune your companions' responses and actions using simple forms.
- Contextual Companions: Create companions that understand and respond based on context, enhancing their utility.
- Configurable Tools: Extend your companion’s capabilities with additional functions, such as internet access, time-based responses, etc.
Planned Features (or already WIP)
- Expanded Functionality: More tools to enhance GPT companion capabilities.
- Skill Development: Pre-built skills for teaching, language practice, coding assistance, and more.
- Vector Search Optimization: Transition vector search from Atlas Cloud to local PostgreSQL for better performance.
Prerequisites
Before you begin, ensure you have met the following requirements:
- Docker
- WhatsApp, Discord and/or Telegram account
- OpenAI account
- MongoDB (Cloud version with a free tier available here)
📝 Installation
To install and run the application locally, follow these steps:
Clone the repository:
git clone https://github.com/catalinberta/panko-gpt.gitNavigate to the project directory:
cd panko-gptRename
.env.exampleto.envRename
.env.dev.exampleto.env.devFill in the missing environment variables (e.g. Atlas credentials, OpenAI key etc.) in
.envand.env.dev.
For Development
Build the development environment:
docker compose build developmentStart the development environment:
docker compose up development
For Production
Build the production environment:
docker compose build productionStart the production environment:
docker compose up production
Running the Application
- Open your browser and go to http://localhost:5005
Community
Join the Discord community to connect with other users, share ideas, and get support.
⭐️ Other links
🤝 Contributing
Contributions are very welcome! Whether it's adding new features, improving documentation, or reporting bugs, please feel free to make a pull request or open an issue.
📃 License
This project is licensed under the MIT License.
Media gallery
1 / 4Install panko-gpt on Unraid in a few clicks.
Find panko-gpt in Community Apps on your Unraid server, review the template, and click Install. Unraid handles the Docker app or plugin setup from the published template.
Categories
Download Statistics
Related apps
Explore more like this
Explore allDetails
catalinbertadev/panko-gpt:latestRuntime arguments
- Web UI
http://[IP]:[PORT:5002]- Network
bridge- Shell
sh- Privileged
- false
Template configuration
- Target
- 5002
- Value
- 5002
You can use the free tier from MongoDB Atlas
IMPORTANT: Ensure the API Keys have 'Organization Owner' permissions (tutorial how to obtain these keys: https://tinyurl.com/25tesjpn)
You can find the Project ID in the url after you log into the atlas cloud account and selecting the project. e.g. https://cloud.mongodb.com/v2/PROJECT_ID_HERE#/overview
e.g. the name of the cluster you set when you first created the MongoDB account
(Optional) if not specified, pankogpt will be used
(Optional) but recommended to help pull certain info like all available models
Container Path: /usr/src/backend/data
- Target
- /usr/src/backend/data
- Default
- /mnt/user/appdata/panko-gpt/data
- Value
- /mnt/user/appdata/panko-gpt/data