game-build
Skill1.1k starsChanged 2 days agoMITView on GitHub
Build the approved game design into a complete, runnable game using the local environment's available rendering stacks, 3D/2D generators, and modeling tools. Turn GAME_DESIGN and ART_DIRECTION into a playable graybox first, then the finished game with rule regressions, one end-to-end verify command, and a concise BUILD_BRIEF. Use for implement the approved game design, prototype the riskiest design question, or turn this design into a running 2D, 2.5D, or 3D game. 游戏构建执行。结合本机与当前运行环境实际可用的渲染栈与生成/建模工具(如 Three.js、图生 3D 模型、Blender、2D/音视频工具),把已批准的 GAME_DESIGN 与 ART_DIRECTION 先做成可玩通的灰盒,再实现为完整可运行的游戏并配齐规则回归与端到端验证命令。用于验证核心设计风险、构建游戏原型或实现正式可玩版本。
1 other project uses a version of this.
Trending
More talk lately.about the whole repository
Most recently taken up by
- fcsouza/agent-skillstheir own version, first seen Nov 14, 2026
Who's talking about zenstory-ai/novel-to-game
- sudosubin/agents.nix“claude-code-plugins.github.zenstory-ai.novel-to-game: init at 0.9.0”GitHub · Nov 16, 2026
- sudosubin/agents.nix“claude-code-marketplaces.github.zenstory-ai.novel-to-game: init at 0.9.0”GitHub · Nov 15, 2026
- sudosubin/agents.nix“agent-skills.github.zenstory-ai.novel-to-game: 0.8.0 → 0.9.0”GitHub · Nov 15, 2026
- zenstory-ai/oh-story-dsh“上游更新:NovelToGame 0.9.0”GitHub · Nov 15, 2026
- zuiyi233/dsh-miaowu“上游更新:NovelToGame 0.9.0”GitHub · Nov 15, 2026
The biggest 5 of 20 mentions linking to the repository that our crawl found. Each line is their own headline.
See all 20
Talked about on GitHub (20 times).
On GitHub · 20
- Nov 16, 2026claude-code-plugins.github.zenstory-ai.novel-to-game: init at 0.9.0 — sudosubin/agents.nix0 reactions
- Nov 15, 2026claude-code-marketplaces.github.zenstory-ai.novel-to-game: init at 0.9.0 — sudosubin/agents.nix0 reactions
- Nov 15, 2026agent-skills.github.zenstory-ai.novel-to-game: 0.8.0 → 0.9.0 — sudosubin/agents.nix0 reactions
- Nov 15, 2026上游更新:NovelToGame 0.9.0 — zenstory-ai/oh-story-dsh0 reactions
- Nov 15, 2026上游更新:NovelToGame 0.9.0 — zuiyi233/dsh-miaowu0 reactions
15 more on GitHub
- Nov 12, 2026agent-skills.github.zenstory-ai.novel-to-game: 0.7.1 → 0.8.0 — sudosubin/agents.nix0 reactions
- Nov 12, 2026上游更新:NovelToGame 0.8.0 — zenstory-ai/oh-story-dsh0 reactions
- Nov 12, 2026上游更新:NovelToGame 0.8.0 — zuiyi233/dsh-miaowu0 reactions · 1 comment
- Nov 8, 2026agent-skills.github.zenstory-ai.novel-to-game: 0.7.0 → 0.7.1 — sudosubin/agents.nix0 reactions
- Nov 8, 2026上游更新:NovelToGame 0.7.1 — zenstory-ai/oh-story-dsh0 reactions
- Nov 8, 2026上游更新:NovelToGame 0.7.1 — zuiyi233/dsh-miaowu0 reactions · 1 comment
- Nov 6, 2026上游更新:NovelToGame 0.7.0 — zuiyi233/dsh-miaowu0 reactions · 1 comment
- Nov 5, 2026agent-skills.github.zenstory-ai.novel-to-game: 0.6.1 → 0.7.0 — sudosubin/agents.nix0 reactions
- Nov 3, 2026[Skill Proposal]: novel-to-game — akhmadilham/pinax-registry0 reactions · 1 comment
- Oct 21, 2026上游更新:NovelToGame 0.6.1 — zenstory-ai/oh-story-dsh0 reactions
- Oct 21, 2026上游更新:NovelToGame 0.6.1 — zuiyi233/dsh-miaowu0 reactions · 1 comment
- Oct 20, 2026agent-skills.github.zenstory-ai.novel-to-game: 0.6.0 → 0.6.1 — sudosubin/agents.nix0 reactions
- Oct 19, 2026agent-skills.github.zenstory-ai.novel-to-game: 0.5.0 → 0.6.0 — sudosubin/agents.nix0 reactions
- Oct 4, 2026agent-skills.github.zenstory-ai.novel-to-game: 0.4.0 → 0.5.0 — sudosubin/agents.nix0 reactions
- Oct 3, 2026agent-skills.github.zenstory-ai.novel-to-game: init at 0.4.0 — sudosubin/agents.nix0 reactions
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- fcsouza/agent-skills★ 26same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository6 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- game-designSkill
- game-qaSkill
- novel-game-analyzeSkill
- novel-to-gameSkill
The file
Files
游戏构建执行
把已批准的 GAME_DESIGN.md 与 ART_DIRECTION.md 实现为目标平台上可完整游玩的游戏。架构、框架、数据结构、渲染方式与测试组织都由你决定,不设模板。灰盒只需要 GAME_DESIGN.md,接入美术之前须有 ART_DIRECTION.md;上游决定缺失或与实现冲突时退回 game-design 修订,不在代码或 BUILD_BRIEF.md 里暗自重做设计。
产物语言由 PRODUCT_BRIEF.md 锁定;未锁定时跟随对话语言,不默认产出中文。
环境与选型
动手前盘点本机与会话实际能调用的运行时、渲染库、命令行、MCP 与生成服务(生图、图生 3D、Blender、视频、音乐、语音),在 PRODUCT_BRIEF.md 批准的平台内选最能兑现作品气质的路线:核心乐趣在空间、手感或镜头上的,就做成真正的 2D、2.5D 或 3D 画面,不压成文字选项。目标工具链缺失时不擅自改做网页版,只走简报已批准的替代环境,并分开记录 targetRuntime 与 testedRuntime;完全无法构建时说明阻塞原因,只交付 BUILD_BRIEF.md。
生成预算优先投向主角、核心地标与关键演出。成品资源保存到工作区并接进真实画面,正式版不留占位图、坏链或没执行的提示词。调用外部工具只传当前资产需要的描述与参考图,不外传原著全文,密钥不进仓库。用到视频过场、生成音乐音效或配音时读 generative-media.md。
构建顺序
- 先灰盒玩通:用最轻的占位把核心循环做到能从初态玩到结局——灰盒同样用真实的空间与操作,不以文字原型代替——交给未读设计与源码的独立会话从头试玩。看不懂、玩不通或核心乐趣不成立就先修(属于设计的问题退回
game-design),不带着这些问题投入美术。targetFinish: graybox时灰盒即交付,跳过第 2 步的美术量产。 - 完整实现并打磨:实现
GAME_DESIGN.md的全部状态、动作、分支与结局,接入美术、演出与声音;在目标视口亲手玩,看实机截图修到位;有批准目标图的界面与目标图并排比对,连续两轮无改善就停手记入限制,不改目标图迁就实现。必需资产加载失败要显式报错;可降级资产切到替代方案后,核心体验仍须成立。 - 审校上屏文字:按 player-copy-polish.md 通查所有玩家看得见的文案。
- 自测与验证入口:为会改变结果的规则、数值门槛、分支汇流与结局条件写回归;提供一条端到端
verify命令,从全新初始状态用真实操作连续跑通启动、渲染、输入、核心循环、到达设计结局与重开回初态(即game-qa的六项检查,记录格式由它规定)。命令支持指定结果输出路径,无论成败都整份重写,不残留旧PASS,也不预填结果;构建自测把自测记录写到qa/之外,正式的qa/verification.json只由game-qa在盲玩之后运行同一条命令写入。 - 写
BUILD_BRIEF.md(骨架见 build-brief-contract.md),移交game-qa做独立盲玩验收。
按需规则
- 接入玩家自由打字输入:模型只把玩家的表述解析成预设的几种标准动作,或把已结算的事实润色成台词;时间、资源、位置、物品、知情、承诺与胜负一律由固定规则判定。解析结果先过前置条件、在场与知情范围、原著硬边界的校验,不通过就追问或拒绝,不因关键词命中把否定句判成相反的行动;模型生成的摘要或好感估值不写入正式状态。回放与校验从标准动作跑起,不重新调用模型;为否定句、缺失对象、篡改原著事实、向不在场角色泄密各写一条回归。
- 存档:节点、状态含义或结算规则变更后,迁移旧档或明确拦截并提示不兼容,不静默读取。
- 连续行走的场景:底图、可走区域、角色脚底锚点、交互热区与小地图共用一套坐标,画面上的道路、桥面、水岸与建筑边界和通行判定逐处对齐,重绘地图后同步校准阻挡、触发点与旧档坐标;寻路从角色当前位置沿合法道路走完,寻路失败时不得直接穿墙或瞬移敷衍过去。
- Blender MCP:配置了不代表连得上,修改前先读一次场景确认;它有本机执行权限,本地端口不暴露公网,不运行来源不明资产包里的脚本。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

