Fish (also known as Literature) is a complex 6 player card game that involves 2 teams, with each team trying to guess the cards of the opposing 3-player team, while simultaneously tracking the cards in their own team. Each player starts with 9 cards, and players are not allowed to know others' cards. Because it requires such a specific amount of players, frequently groups find themselves underpopulated and unable to play properly. We built this bot with the aim of replacing player(s) so that smaller groups can also enjoy this card game.

What it does

Play Fish. Specifically, it can perform all of the major actions a human player would do: ask for cards, relinquish cards, and making claims.

How we built it

FishGo was built in Python. We are in the process of implementing Google's Text-to-Speech API.

Challenges we ran into

Implementing the many more complex strategies that are used in Fish by human players, such as avoiding a certain opponent or asking for a specific suite of cards to provide more information to teammates.

Accomplishments that we're proud of

Successfully creating a bot that can follow the rules of the game and track cards, meaning it can already, at its current stage, replace a human player.

What we learned


What's next for FishGo

Implement machine learning so that the bot can develop its own strategies for the game.

Built With

Share this project: