Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add alfredxw/denova --skill game-craftgit clone --depth 1 https://github.com/alfredxw/denovaWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/alfredxw/denova/game-craft)<a href="https://agentmods.dev/skills/alfredxw/denova/game-craft"><img src="https://agentmods.dev/badge/skills/alfredxw/denova/game-craft/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/alfredxw/denova/game-craft"><img src="https://agentmods.dev/badge/skills/alfredxw/denova/game-craft.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00092 | $0.03119 |
| Opus 5.5 | $0.00037 | $0.01248 |
| Sonnet 5.5 | $0.00018 | $0.00624 |
| Haiku 4.5 | $0.00009 | $0.00312 |
Grade A, and why
game-craft scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
互动游戏创作
把游戏设计判断落成作者能直接使用的交付物。原则用于取舍，不用于逐项打分；作者的题材、玩法、预期体验和已确认设定优先。
本技能是创作判断层，只补充判断、模板和诊断，不决定执行流程。游戏前期设计、资料库、状态/规则/规划模板和开局文本开始前加载；读取工作区、写入资料、提交配置、跑回合和状态同步仍按当前流程和本次授权执行。本技能第一版面向创作 Agent，不替代 Game Agent 的回合协议；游戏引擎或插件开发不属于本技能，使用 extension-development。
开工前先确认三件事
- 玩家是谁、来做什么、为什么留下：玩家扮演谁？反复做什么？什么让他愿意再来一回合？
- 哪些是已确认事实，哪些只是可能：canon、当前状态、未来计划分别是什么？哪些必须留给玩家决定？
- 这一局要记住什么：哪些必须跨回合稳定？哪些只是即兴？如果只保留三个状态值，是哪三个？
六条核心判断
| 原则 | 操作问题 | 常见失效 |
|---|---|---|
| 让玩家做有代价的选择 | 每个选项失去什么、得到什么、改变了谁？ | 选项只有措辞不同 |
| 让世界在玩家之外运转 | 玩家不动时，谁按什么计划行动？ | NPC 原地等待，世界静止 |
| 让后果改变局势 | 成功/失败后，位置、资源、关系、信息变了什么？ | 判定完回到原状 |
| 让规则制造取舍 | 这条规则让哪种做法更划算？ | 为一切掷骰，或掷骰不改变选择 |
| 让资料可玩 | 这条设定能引发什么行动、冲突或代价？ | 百科式设定，玩家无法使用 |
| 让状态与计划只保留会影响下一步的部分 | 这个值下次何时被读取？这个计划玩家偏航后还能用吗？ | 面板越做越大，计划变成剧本 |
网文向游戏还要额外检查：成长是否可感、爽点是否有铺垫、金手指是否有代价、世界是否有阶梯、日常是否有钩子。经典网文的结构是参照，不是必须凑齐的清单；先确认玩家体验，再决定借用哪些。成长、爽点、资源、世界阶梯、题材 lens 见 web-novel-patterns.md。
三层账本 + 两个契约
| 层 | 是否必须 | 回答的问题 | 放什么 | 不放什么 |
|---|---|---|---|---|
| 资料库（Lore） | 必须 | 世界长期成立的事实是什么？谁能做什么？代价是什么？ | 人物、地点、势力、物品、世界规则、历史、秘密的真相 | 当前伤势/位置/临时目标、未来剧情、玩家未确认的猜测 |
| 状态系统（State） | 可选 | 这一局现在怎样？什么会变？玩家能看到什么？ | 位置、时间、资源、关系、时钟、旗标、已知信息、伤势 | 完整人物小传、稳定设定、未来计划 |
| 游戏规划（Planning） | 可选 | 接下来可能怎样？谁在行动？何时揭示？ | 长线方向、阶段、NPC 议程、线索/回收、候选情境、分支可能 | 已完成回合、精确状态值、强制玩家选择 |
| 规则与判定（Rules） | 可选 | 不确定性如何变成后果？ | 触发条件、赌注、难度依据、结果带、失败推进、状态绑定 | 世界 canon、人物小传、文风要求 |
| 开局与呈现（Opening / Presentation） | 必须 | 玩家如何进入？第一眼看到什么？每回合如何获得反馈？ | 开场情境、第一个选择、文风、长度、视角、内容边界、状态展示 | 大段设定说明、未来剧情、第二个状态系统 |
两个契约：
- 规则契约：世界法则放 Lore，判定规则放 Rule System，叙述规则放 Narrative Style。不要用世界法则代替判定，也不要把判定写进设定。
- 信息契约：真相放 Lore，玩家已知放 State，揭示时机放 Planning。不要把玩家、角色和导演的知识边界混在一起；也不要把导演真相塞进状态账本。
场景路由
根据作者的任务选择对应参考，只读相关文档；跨场景时再补读。这些是同一技能的参考文档，无须调用其他技能或依次完成所有场景。
参考位置：skill://game-craft/references/world-and-cast.md、skill://game-craft/references/character-personality.md、skill://game-craft/references/state-and-rules.md、skill://game-craft/references/planning-and-scenes.md、skill://game-craft/references/opening-and-diagnosis.md、skill://game-craft/references/web-novel-patterns.md。
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today First seen · 127 lines · 92 tokens per session scan A 81f6f998e043
game-craft is a skill published in the GitHub repository alfredxw/denova (961 stars, last pushed today), licensed Apache-2.0. It adds 92 tokens to every session and 3,119 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-22.
Other skills, from other repositories
novel-workflow-06-emulation-tick
Instructions for advancing an older simulation or role-playing project one world tick at a time. A world tick is one update of the simulated world, such as a turn or time step.
avalon-8-10p
Eight-to-ten-player Avalon rules for a permanent Agent host, with the user choosing to join the randomized player pool or spectate.
clay
AI 3D model generation agent. Generates text-to-3D and image-to-3D code (Python/JS/OpenSCAD) using Meshy, Tripo, Hunyuan3D, Rodin, Sloyd, and Stability APIs. Handles game pipeline integration, LOD, retopology, UV, and QC validation.
llmlint
A Chinese-prose rule library with two entry points. Before writing, emit the rules as writing constraints to load into a system prompt or style guide. After writing, lint the draft with static rule hits, neural AIGC heatmaps, contextual review, approved repair, and local learning notes. Use when drafting or rewriting…
profile-system-guide
Guide users and agents through Neuro Book harness, TSX profiles, ProfileTurnPlan, skills, user-assets overlays, settings forms, profile home resources, profile compile checks, templates, system restore, and safe profile editing.
novel-technique-character-card-workshop
A workshop for turning rough character ideas into detailed, usable character profiles. It supports structured question cycles, personality layers, visual themes, and checks against reducing a character to a few labels.