Project bootstrap
Skill lzx-Bill/Project-Factory-Core/.agents/skills/project-bootstrap
Use when starting a new Project Factory project. Trigger when user says "new project", "create project", "initialize", "start a new wiki", or mentions setting up a fresh project folder with HOME.md and wiki structure. Also use when scaffolding prompts/, acceptance/, tasks/, reports/ directories for an existing project.From its SKILL.md
npx -y skills add lzx-Bill/Project-Factory-Core --skill project-bootstrapAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 25 days oldThe repository was created 25 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
3.4 KB, ~1.1k tokens by cl100k_base, as published. Nobody here has run it
Project Bootstrap
用于新项目启动和初始骨架建立。
When to Use
- 用户刚提出一个新项目想法，需要建立孵化空间
- 需要创建
HOME.md和基础 wiki 目录结构 - 需要初始化
prompts/、acceptance/、tasks/、reports/等交付目录
Input
| 来源 | 内容 |
|---|---|
| 用户消息 | 项目名称（如有）、一句话想法、项目背景 |
| 已有文档 | 无（全新项目） |
Output Schema
| 文件 | 类型 | 说明 |
|---|---|---|
<项目根>/HOME.md | 导航首页 | 一句话定义 + 阶段 + 目标 + 下一步 + 导航 |
<项目根>/wiki/00-overview/vision.md | 愿景页 | 问题、用户、价值、非目标、成功标准 |
<项目根>/wiki/01-requirements/user-stories.md | 用户故事 | 核心场景的 2-3 条用户故事 |
<项目根>/wiki/01-requirements/assumptions.md | 假设页 | 初始假设清单（可占位） |
Minimum Viable Output
- HOME.md 含一句话定义、当前阶段（"启动中"）、当前目标、下一步
- 至少 1 个 wiki 页面有实质内容（不全是占位符）
- 目录结构完整，可继续扩写
Complete Output
- HOME.md + vision.md + user-stories.md + assumptions.md 均含实质内容
prompts/、acceptance/、tasks/、reports/目录已创建tasks/backlog.md含初始任务候选项
Dependencies
| 类型 | 说明 |
|---|---|
| 前置 | 无（从零开始） |
| 后置 | incubation-discovery（方向发散）、overview-framing（愿景细化）、requirements-spec（需求细化） |
| 并行 | 无 |
Procedure
- 确认项目名称，固定项目根目录为
<项目根目录>/ - 创建最小可运行目录结构，所有初始化文件都放在
<项目根目录>/下 - 优先写 HOME.md（一句话定义 + 阶段 + 目标 + 下一步 + 导航）
- 再补 vision.md 和 user-stories.md 初稿
- 不在启动阶段过早写满全部细节页
Enriched Behavior
- 不只创建空目录，至少落下能让用户继续阅读和推进的首批内容
- 首页优先给出一句话定义、当前阶段、当前目标、下一步，而不是只放目录树
- 如果用户描述仍然模糊，用占位性的候选方向和待确认项承接，不要因为信息不全而停止产出
- 优先让新项目具备"可继续扩写"的骨架，而不是一次写满所有页面
Target Pages
<项目根目录>/HOME.md<项目根目录>/wiki/00-overview/vision.md<项目根目录>/wiki/01-requirements/user-stories.md<项目根目录>/wiki/01-requirements/assumptions.md<项目根目录>/prompts/<项目根目录>/acceptance/<项目根目录>/tasks/<项目根目录>/reports/
Changelog
| 日期 | 变更 | 原因 |
|---|---|---|
| 2026-04-26 | 增强：新增 Input/Output Schema、Dependencies、Min/Complete Output 两级标准、Changelog | 统一 skill 结构，提升触发准确性和输出一致性 |
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.
Gives 0 of the 12 instructions most project setup skills give in ~1.1k tokens
Counted across 999 of the 1,637 authors here whose files we hold, read 2026-08-07
- Ask one question at a timein 29 of 999, across 28 files
- Detect the package manager from lockfilesin 28 of 999, across 9 files
- Present findings to the userin 26 of 999, across 5 files
- Explore current repo statein 24 of 999, across 3 files
- Update the agent skills block in place if it existsin 24 of 999, across 3 files
- Install husky lint-staged and prettierin 23 of 999, across 4 files
- Create the lintstagedrc filein 22 of 999, across 3 files
- Commit all changed filesin 22 of 999, across 3 files
- Run lint-staged to verify it worksin 22 of 999, across 3 files
- Create the husky pre-commit filein 21 of 999, across 2 files
- Create a prettierrc file if missingin 21 of 999, across 2 files
- Initialize huskyin 21 of 999, across 2 files
Said here and by no other author read
- write HOME.md first
- draft vision and user-stories pages
- create delivery directories
- seed backlog with candidate tasks
- use placeholders if user details are vague
- ensure skeleton is expandable
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.