The Scripts of Tribute Python library is a wrapper for creating AI bots to compete in the Scripts of Tribute programming game. It facilitates communication with the game's C# .NET engine via gRPC, allowing you to implement custom bots and simulate game scenarios.
To install the library, use pip
:
pip install scripts-of-tribute
To create your own bot, you need to inherit from the ScriptsOfTribute.base_ai.BaseAI
class and implement the required methods:
def pregame_prepare(self):
"""Optional: Prepare your bot before the game starts."""
pass
def select_patron(self, available_patrons: List[PatronId]) -> PatronId:
"""Choose a patron from the available list."""
raise NotImplementedError
def play(self, game_state: GameState, possible_moves: List[BasicMove], remaining_time: int) -> BasicMove:
"""Choose a move based on the current game state."""
raise NotImplementedError
def game_end(self, final_state):
"""Optional: Handle end-of-game logic."""
pass
What's important here in the play
method that bot should return BasicMove
object from the list, it is because Move
objects come from the engine with an Identification number move_id
which is used to quickly identify whether move is legal or not.
The ScriptsOfTribute.game.Game
class is used to register and run your bots. Here's how to use it:
from ScriptsOfTribute.game import Game
from Bots.RandomBot import RandomBot
from Bots.MaxPrestigeBot import MaxPrestigeBot
def main():
bot1 = RandomBot(bot_name="RandomBot")
bot2 = MaxPrestigeBot(bot_name="MaxPrestigeBot")
game = Game()
game.register_bot(bot1)
game.register_bot(bot2)
game.run(
"RandomBot",
"MaxPrestigeBot",
start_game_runner=True,
runs=10,
threads=1,
)
if __name__ == "__main__":
main()
Game.run Parameters:
- bot1Name: Name of the first bot.
- bot2Name: Name of the second bot.
- start_game_runner: If True, the game runner starts automatically. Set to False if you want to run the engine manually.
- runs: Number of games to run.
- threads: Number of threads to use for parallel execution.
- enable_logs: Logging level ("NONE", "INFO", "DEBUG", etc.).
- log_destination: File path to save logs.
- seed: Random seed for reproducibility.
- timeout: Timeout for each game (in seconds).
- base_client_port: Base port for gRPC client communication (default: 50000).
- base_server_port: Base port for gRPC server communication (default: 49000).
The Game
class manages bot registration, game execution, and cleanup.
Key Methods:
- register_bot: Register a bot instance.
- run: Start the game with the specified bots and parameters.
- _cleanup_processes: Clean up all running processes on exit.
Port Assignment:
- The library assigns ports incrementally for multiple threads or bots:
- First bot: 50000 (client), 49000 (server).
- Second bot: 50001 (client), 49001 (server).
- Additional threads increment ports accordingly.
On the first run with start_game_runner=True
the Game
class will download newest GameRunner suited for your operating system from our official releases page, unzip it and then run it. After that if Game
class find the GameRunner it won't download it again.
Here’s a complete example of creating and running two bots:
from Bots.RandomBot import RandomBot
from Bots.MaxPrestigeBot import MaxPrestigeBot
from ScriptsOfTribute.game import Game
def main():
bot1 = RandomBot(bot_name="RandomBot")
bot2 = MaxPrestigeBot(bot_name="MaxPrestigeBot")
game = Game()
game.register_bot(bot1)
game.register_bot(bot2)
game.run(
"RandomBot",
"MaxPrestigeBot",
start_game_runner=True,
runs=10,
threads=1,
)
if __name__ == "__main__":
main()
This code is available in the examples
directory, as well with the example bots.
if you would like to work with the code locally you might need to (re)generate protobuf
files.
The library uses gRPC for communication with the C# .NET engine. The .proto
files are located in the ScriptsOfTribute/Protos
folder. To generate the necessary Python files, run:
python -m grpc_tools.protoc -IProtos --python_out=./Protos/ --grpc_python_out=Protos/. Protos/enums.proto Protos/basics.proto Protos/main.proto
This will generate the required gRPC Python files in the Protos
folder.
This project is licensed under the MIT License. See the LICENSE file for details.
Happy bot building! 🚀