About this project
ARLQ (Another Rogue-Like Quest) is a roguelike game developed collaboratively between human developers and ChatGPT. The codebase was initially generated by ChatGPT based on high-level descriptions, then refined by humans. ChatGPT also contributed to documentation, algorithm design (like maze generation), library usage explanations, type hints, comments, and even monster design ideas. The manual was drafted by humans and polished with ChatGPT’s help.
The game features procedurally generated dungeons where players control a character (@) using arrow keys. A fog-of-war system hides unexplored areas; visited tiles remain visible via auto-mapping. Monsters (initially shown as '?') and companions ('!') reveal their identities upon encounter. Monsters don’t move unless approached; defeating them grants experience, items, or LP restoration. Some monsters have rare variants with special effects (e.g., Rare Amoeba boosts level significantly, Rare Chimera gives a cursed sword).
Players must manage LP (Life Points), which deplete with each move. Defeating monsters restores food and LP. Companions provide passive benefits: Nomicon reveals monster types, Ocular extends vision, Pegasus helps bypass walls. The goal is to find treasure chests guarded by dragons — Stage 1 by Dragon (D), Stage 2 by Fire Drake (F). Game ends when player touches the treasure chest (T).
Installation uses pipx: `pipx install git+https://github.com/tos-kamiya/arlq`. Options include `-T` (easier view), `-F` or `-t` (harder), and `--curses` for terminal mode (requires platform-specific curses package). The project is BSD-2 licensed.
Comments
0 Rating appears after 10 ratings
Sign in to join the discussion.