npx skills add https://github.com/jwynia/agent-skills --skill list-builder你为创意随机化构建全面、高质量的列表。这些列表将输入到熵工具中,为故事发展注入不可预测性。
好的熵列表具有三个属性:
LLMs 擅长研究、分类和质量控制。脚本擅长存储和随机选择。此技能在两者之间架起桥梁。
完整标准请参见 references/dataset-quality-criteria.md。
| 级别 | 数量 | 状态 | 用例 |
|---|---|---|---|
| 入门级 | 10-30 | 快速示例 | 原型设计、演示 |
| 功能级 | 30-75 | 可用但有限 | 个人项目 |
| 生产级 |
广告位招租
在这里展示您的产品或服务
触达数万 AI 开发者,精准高效
| 75-150 |
| 可常规使用 |
| 客户工作、已发布的工具 |
| 全面级 | 150+ | 参考质量 | 权威资源 |
关键指标:
当前内置列表为入门级/功能级。 此技能旨在将它们提升至生产级。
好: "电梯检查员"(具体、出乎意料、引发问题) 差: "办公室职员"(通用、预期、无亮点)
好: "午夜的自助仓储设施"(具体时间、暗示氛围) 差: "建筑物"(过于模糊,无法使用)
好: "他们正在侦破一个完全不同的案件,但使用了相同的证据"(具体的冲突机制) 差: "他们碍事了"(无机制,只有效果)
构建列表时,确保覆盖相关维度:
职业:
地点:
角色特质:
从立刻想到的 10-20 个项开始。这些是"可用"选项——任何人都会想到的那些。它们有效但不够充分。
使用可用资源扩展到显而易见之外:
Kiwix/Wikipedia:
模式:维度扩展
移除以下项:
输出为 JSON 数组,供 entropy.ts 使用:
{
"list_name": [
"Item one",
"Item two",
"Item three"
]
}
分析列表的质量和多样性。
deno run --allow-read scripts/validate-list.ts list.json
# 检查文件中的特定列表
deno run --allow-read scripts/validate-list.ts data.json professions
报告内容:
合并多个列表源,去重并格式化。
deno run --allow-read scripts/merge-lists.ts source1.json source2.json --output combined.json
当你需要研究特定类别时,使用如下提示:
对于职业: "找出 [行业] 中 20 个大多数人不知道其存在的职业。专注于涉及有趣权限、专业知识或异常工作条件的工作。"
对于地点: "找出 20 个可能发生重要对话的具体地点(而非类别)。专注于具有内在张力、时间压力或意外亲密感的地方。"
对于角色缺陷: "找出 20 个人对自己持有的具体错误信念,这些信念并非明显的反派特质。专注于那些感觉具有保护性但实际上具有限制性的信念。"
Kiwix 搜索:"List of occupations" → 分类页面 → 具体的不寻常工作
从研究中添加:
移除:
使用此技能构建的列表将放入:
story-sense/data/ 用于小说特定列表entropy.ts --file 加载命名约定:
[类别]-[具体性].jsonprofessions-unusual.json、locations-liminal.json、objects-evidence.json此技能将主要输出写入文件,以便工作在不同会话间持久保存。
在进行任何其他工作之前:
context/output-config.mddata/ 或 story-sense/data/ 用于熵列表context/output-config.md 中.list-builder-output.md 中对于此技能,持久化:
| 存入文件 | 保留在对话中 |
|---|---|
| 最终列表(JSON) | 关于列表用途的讨论 |
| 研究来源 | 项的迭代 |
| 质量分析 | 实时反馈 |
| 文档 | 类别细化 |
模式:{类别}-{具体性}.json 示例:professions-unusual.json
每周安装量
91
代码库
GitHub 星标数
42
首次出现
Jan 20, 2026
安全审计
安装于
codex79
opencode79
gemini-cli78
cursor75
github-copilot73
cline66
You build comprehensive, high-quality lists for creative randomization. These lists feed into entropy tools that inject unpredictability into story development.
Good entropy lists have three properties:
LLMs are good at research, categorization, and quality control. Scripts are good at storage and random selection. This skill bridges them.
See references/dataset-quality-criteria.md for complete criteria.
| Level | Size | Status | Use Case |
|---|---|---|---|
| Starter | 10-30 | Quick example | Prototyping, demos |
| Functional | 30-75 | Usable but limited | Personal projects |
| Production | 75-150 | Ready for regular use | Client work, published tools |
| Comprehensive | 150+ | Reference quality | Definitive resource |
Key metrics:
Current built-in lists are Starter/Functional level. This skill exists to build them up to Production.
Good: "Elevator inspector" (specific, unexpected, sparks questions) Bad: "Office worker" (generic, expected, no hooks)
Good: "Self-storage facility at midnight" (specific time, atmosphere implied) Bad: "Building" (too vague to use)
Good: "They're solving a completely different case that uses same evidence" (specific collision mechanism) Bad: "They get in the way" (no mechanism, just effect)
When building a list, ensure coverage across relevant dimensions:
Professions:
Locations:
Character traits:
Start with 10-20 items that come to mind immediately. These are the "available" options—the ones that would occur to anyone. They're valid but not sufficient.
Use available sources to expand beyond obvious:
Kiwix/Wikipedia:
Pattern: Dimensional expansion
Remove items that are:
Output as JSON array for use with entropy.ts:
{
"list_name": [
"Item one",
"Item two",
"Item three"
]
}
Analyzes a list for quality and variety.
deno run --allow-read scripts/validate-list.ts list.json
# Check specific list in a file
deno run --allow-read scripts/validate-list.ts data.json professions
Reports:
Combines multiple list sources, deduplicates, and formats.
deno run --allow-read scripts/merge-lists.ts source1.json source2.json --output combined.json
When you need to research a specific category, use prompts like:
For professions: "Find 20 professions in [industry] that most people don't know exist. Focus on jobs that involve interesting access, specialized knowledge, or unusual working conditions."
For locations: "Find 20 specific locations (not categories) where important conversations might happen. Focus on places with built-in tension, time pressure, or unexpected intimacy."
For character flaws: "Find 20 specific false beliefs people hold about themselves that aren't obvious villain traits. Focus on beliefs that feel protective but are actually limiting."
Kiwix search: "List of occupations" → Category pages → specific unusual jobs
Add from research:
Remove:
Lists built with this skill go into:
story-sense/data/ for fiction-specific listsentropy.ts --fileNaming convention:
[category]-[specificity].jsonprofessions-unusual.json, locations-liminal.json, objects-evidence.jsonThis skill writes primary output to files so work persists across sessions.
Before doing any other work:
context/output-config.md in the projectdata/ or story-sense/data/ for entropy listscontext/output-config.md if context network exists.list-builder-output.md at project root otherwiseFor this skill, persist:
| Goes to File | Stays in Conversation |
|---|---|
| Final list (JSON) | Discussion of list purpose |
| Research sources | Iteration on items |
| Quality analysis | Real-time feedback |
| Documentation | Category refinement |
Pattern: {category}-{specificity}.json Example: professions-unusual.json
Weekly Installs
91
Repository
GitHub Stars
42
First Seen
Jan 20, 2026
Security Audits
Gen Agent Trust HubPassSocketPassSnykWarn
Installed on
codex79
opencode79
gemini-cli78
cursor75
github-copilot73
cline66
AI 代码实施计划编写技能 | 自动化开发任务分解与 TDD 流程规划工具
50,900 周安装
阿里云CDN OpenAPI自动化操作指南 - 域名管理、缓存刷新、HTTPS证书配置
129 周安装
pnpm 完全指南:JavaScript/TypeScript 包管理器安装、配置与最佳实践
134 周安装
FAISS 向量相似性搜索库 - Meta AI 十亿级向量快速检索,支持 GPU 加速与 Python 集成
130 周安装
Figma设计插件开发与组件系统指南:从自动布局到设计系统管理
130 周安装
tuzi-post-to-x:绕过反机器人检测,自动化发布内容到X/Twitter
70 周安装
MLflow 机器学习生命周期管理平台:实验跟踪、模型注册与部署
131 周安装