$ show trevorbayless/cli-chess
cli-chess is a terminal-based chess application that supports playing online via Lichess.org and offline against the Fairy-Stockfish engine. It features a modular MVP (Model-View-Presenter) architecture with clear separation of concerns across core game logic, reusable UI components (board, move list, chat, clock), menu systems, and API integration. The application supports all Lichess chess variants, includes features like premoves, takebacks, draw offers, chat, blindfold mode, unicode pieces, custom theming, and PGN export. Entry point is src/cli_chess/__main__.py, with gameplay driven by user input flowing through presenters to models, which update views via an event system. External dependencies include python-chess for board logic, berserk for Lichess API, and prompt-toolkit for the terminal UI.
The LLM identified 3 files worth highlighting for reuse:
Implements a lightweight observer pattern for decoupling components. Provides Event and EventManager classes that allow publish/subscribe communication between models, presenters, and views without tight coupling. Self-contained with zero internal dependencies, making it easily extractable to other projects. Demonstrates clean implementation of pub/sub with proper listener management and notification handling.
Handles application configuration persistence across Windows, Linux, and macOS platforms. Manages multiple config sections (player_info, game, terminal, lichess) with type conversion, default values, and change notifications. Self-contained with clear separation of concerns, allowing extraction to other CLI applications. Shows thoughtful design for cross-platform config paths, automatic section recreation, and event-driven updates.
Contains reusable utility decorators and functions including @threaded for running functions in daemon threads and @retry for flexible retry mechanisms with configurable attempts and exception types. Also includes cross-platform OS detection, string-to-boolean conversion, and URL opening utilities. These utilities are self-contained, demonstrate solid decorator patterns, and have minimal dependencies, making them highly reusable in other Python projects.