Home

Pan's Trial Wiki

Pan's Trial is a two-player tactical card-and-labyrinth game built for ECE 348. Players draft high-value cards, reveal Omens that assign each suit a changing role, move through a 6x6 wraparound labyrinth, fight with weapon cards, and enter Appeasing Pan to reshape the board or health state.

This wiki is a GitHub-ready documentation bundle for the current project. It is meant to support players, judges, teammates, and future contributors.

Start Here

• Getting Started
• Gameplay and Rules
• Project Architecture
• Multiplayer and Rooms
• Web Build and Deployment
• Testing and Balance
• Development Guide
• Roadmap
• Publishing the Wiki

Project Snapshot

• The full gameplay loop is implemented: coin flip, draft, Omen reveal, Traversing, Appeasing Pan, Pan requests, post-Appeasing hole placement, victory resolution, match summary, and rematch flow.
• The project supports single player against AI, two-player room-code multiplayer, desktop play from source, a Windows executable, and a browser build.
• The codebase is split into a headless rules engine, a Pygame UI, multiplayer transport layers, browser build tooling, and automated regression tests.
• The same engine powers gameplay, tests, multiplayer state, and headless balance simulations.

Suggested Reading Paths

• New players: start with Getting Started, then read Gameplay and Rules.
• Developers: start with Project Architecture, then use Development Guide and Testing and Balance.
• Deployment and demo prep: start with Web Build and Deployment, then read Multiplayer and Rooms.

Core Files and Folders

• main.py starts the desktop and browser game.
• engine/ contains the headless game rules.
• ui/ contains screens, rendering, input, and audio.
• multiplayer/ contains room clients, room state helpers, and serialization.
• room_server.py hosts the room API and can also serve the web build.
• build_web.py creates the browser package and deployment site.
• tests/test_rules.py contains the main regression suite.
• balance_testing.py runs AI-vs-AI balance studies without the UI.