Claude Code wrote a multiplayer wizard game in 19 days: 1,000 commits, 106k lines
Difficult_Ocelot_660 · reddit · 2026-10-21
A developer shipped Spellblunder, a browser-based multiplayer 3D wizard dueling game (Magicka-like element mixing), in 19 days with all code written by Claude Code (Opus 5.5); his role was ideas and heavy playtesting.
Key numbers: 1,000 commits, 106k lines of TypeScript for the game, 6k for the server, 59k for tests, 2,552 automated tests. Ships with a five-chapter campaign, arena, online co-op/duels, bots and six languages.
Notable workflow details:
- Multiple parallel Claude Code sessions, each in its own git worktree, merged one by one
- Claude made three mockups per new screen; he picked one
- Online play is lockstep (every browser must compute identical state) — Claude built a determinism bench that replays scripted matches and hashes results across Node and real browsers
- Pre-launch multi-agent bug hunt: finders, two verifiers per bug, one fixer per file group — 92 bugs fixed
- Natural-language audio pipeline: one prompt had Claude generate 8 ElevenLabs footstep variants, trim them, match loudness and stage them on a listening bench
What Claude couldn't do: judge whether a fight feels fair or a spell feels good — that stayed human.
More from coding & agent
- Claude cut a 2.6-hour video export to 20 minutes — but only after being told to rebuild the server — byhewar · 2026-10-23
- 3 flight-booking rules my AI agent checks that I'd never do by hand — Efistoffeles · 2026-10-23
- OpenJet claims Qwen3.8-27B coding runs 1.9x faster than llama.cpp on one RTX 3090 — JeremyCMorgan · 2026-10-23
- Building an autonomous SEO agent in Hermes with three skills: programmatic pages, Keywords Everywhere, DataForSEO — EXM7777 · 2026-10-23
- Building a sports betting AI agent with onchain execution and human approval by default: architecture and lessons — Dry-Comfortable-2678 · 2026-10-23
- How do you audit AI agent decisions in regulated workflows? A field survey — Ok_Way_1993 · 2026-10-23