sanhuang520-ship-it/awesome-chinese-ai-tools / skills/github-readme-cn
github-readme-cn
Verified中文开源项目的 README 与仓库门面优化。当用户说"我的开源项目没人看""README 怎么写""GitHub 仓库怎么优化""帮我写 README""项目主页太丑""怎么让别人愿意 star"时使用。基于对 15 个近期高增长仓库的实测数据给出可执行建议,同时明确说明哪些是相关性、哪些无法验证——不承诺涨星。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 7.2 KB |
| references/measured-data.md | markdown | 6.5 KB |
references/measured-data.md
View on GitHub# 实测数据:15 个高增长仓库的仓库门面
> 采集时间 2026-08-07,用 GitHub REST API。
> **样本是"已经火了的",存在幸存者偏差,下面所有数字都只是相关性。**
## 样本与筛选
从 GitHub 搜索近 60–180 天创建、星数较高的仓库中取 15 个,
覆盖英文/中文、工具/列表/Skill 三类。**没有对照组**——
用同样结构但没火的仓库拿不到(GitHub 搜索按星排序,低星的搜不到)。
这是这份数据最大的局限。
## 一、结构特征
| 仓库 | 星/天 | README | 图 | 首屏图 | 徽章 | 首屏demo | 首屏装 |
|------|------:|-------:|---:|------:|:----:|:-------:|:-----:|
| DietrichGebert/ponytail | 1751 | 19KB | 13 | 7 | ✅ | | |
| KKKKhazix/human-writing | 920 | 5KB | 4 | 4 | ✅ | | ✅ |
| img2threejs/img2threejs | 441 | 26KB | 16 | 9 | ✅ | | |
| thebuggeddev/anatomy | 389 | 3KB | 0 | 0 | | | ✅ |
| mshumer/Claude-of-Duty | 227 | 6KB | 0 | 0 | | | ✅ |
| ayghri/i-have-adhd | 210 | 3KB | 2 | 2 | ✅ | | |
| Donchitos/Claude-Code-Game-Studios | 135 | 15KB | 10 | 7 | ✅ | | |
| helloianneo/ian-xiaohei-illustrations | 127 | 8KB | 9 | 0 | | | |
| freestylefly/awesome-gpt-image-2 | 95 | 32KB | 49 | 6 | ✅ | | |
| laoma2053/awesome-zhuiju-free | 86 | 42KB | 30 | 4 | ✅ | ✅ | |
| isjiamu/gzh-design-skill | 78 | 19KB | 20 | 5 | ✅ | | ✅ |
| vinhhien112/img2obj | 54 | 7KB | 4 | 2 | | ✅ | |
| jnMetaCode/superpowers-zh | 54 | 26KB | 10 | 5 | ✅ | | |
| agiwhitelist/auteur | 33 | 14KB | 17 | 6 | ✅ | ✅ | |
| VoltAgent/awesome-claude-design | 30 | 15KB | 8 | 5 | ✅ | | |
**汇总**
- 首屏有图 **12/15**
- 首屏有徽章 **11/15**
- 首屏有安装命令 4/15
- 首屏能点到 demo 3/15
- 标题用中文 2/15
- README 中位数 **15 KB**,图片中位数 **10 张**
- 仓库名长度中位数 **14 字符**,topics 中位数 **9 个**
**一个反直觉的点**:`VoltAgent/awesome-claude-design` 全仓库只有 **2 个文件、16 KB**,
拿到 3,349 星。体量和结果没有明显关系,**定位清不清楚更重要**。
## 二、作者影响力(判断"打法可复制性"的关键)
| 仓库 | 星 | 作者粉丝 | 账号年龄 | 公开库 |
|------|---:|--------:|--------:|------:|
| DietrichGebert/ponytail | 98,055 | 1,414 | 3年 | 3 |
| Donchitos/Claude-Code-Game-Studios | 23,684 | 262 | 2年 | 1 |
| ayghri/i-have-adhd | 18,031 | 183 | 8年 | 78 |
| img2threejs/img2threejs | 10,150 | 64 | 0年 | 4 |
| freestylefly/awesome-gpt-image-2 | 9,855 | 1,479 | 7年 | 81 |
| helloianneo/ian-xiaohei-illustrations | 9,150 | 488 | 4年 | 6 |
| jnMetaCode/superpowers-zh | 7,541 | 715 | 11年 | 122 |
| **laoma2053/awesome-zhuiju-free** | **5,438** | **47** | 2年 | 23 |
| VoltAgent/awesome-claude-design | 3,349 | 3,774 | 1年 | 19 |
| mshumer/Claude-of-Duty | 2,949 | 1,730 | 8年 | 41 |
| isjiamu/gzh-design-skill | 2,901 | 73 | 9年 | 32 |
| thebuggeddev/anatomy | 1,944 | 512 | 0年 | 71 |
| KKKKhazix/human-writing | 1,841 | 2,963 | 0年 | 3 |
| **vinhhien112/img2obj** | **1,566** | **14** | 8年 | 4 |
| **agiwhitelist/auteur** | **901** | **4** | 0年 | 10 |
**分布**
```
0–50 粉(素人) 3 个 ███
50–500 粉 5 个 █████
500–5000 粉 7 个 ███████
5000+ 粉(大 V) 0 个
```
**读法**:
- 没有一个是大 V,说明这不是"名人效应"游戏
- 但 12/15 有 50 粉以上,说明"完全零基础"是少数(3/15 = 20%)
- 最鼓舞人的样本是 `laoma2053`:**47 个粉丝,做到 5,438 星**
## 三、测不到的(重要)
**stargazer 时间线全部 404。**
尝试用 `/repos/{owner}/{repo}/stargazers` 配 `application/vnd.github.star+json`
拉前 100 个 star 的时间戳,想据此判断是"一夜爆发"(≈外部推荐)还是"持续增长"(≈内容驱动)。
对自己的仓库可以,对别人的仓库**一律 404**。
**后果**:无法区分某个仓库是靠 README 好,还是靠某个大号转发。
这是"打法"最核心的因果问题,而这份数据回答不了。
## 四、中文搜索词在 GitHub 上基本不工作(2026-08-28 补测)
给两个中文项目挑 topic 时顺带测的,结论比 topic 清单本身有用。
用 `/search/repositories` 分别以中英文查询同一主题,取按星排序的前 15:
| 查询词 | 前 5 名的相关性 |
|---|---|
| `元素周期表` | **0 / 5 相关**。返回的是某书单仓库(★22871)、x86 汇编示例、某政治仓库、某组织主页 |
| `互动学习 物理 化学` | **0 个结果** |
| `教学 可视化` | **0 / 5 相关** |
| `periodic table interactive` | **5 / 5 相关**,全是真的周期表项目 |
| `interactive science education` | 4 / 5 相关 |
| `chemistry visualization` | 5 / 5 相关 |
**推论**:想靠搜索被找到,topic 和描述里的关键词要用英文。
中文项目当然可以有中文说明,但**别指望中文关键词把人带过来**。
### 顺带得到的 topic 分布
统计上述查询命中的仓库实际打了哪些 topic:
- 化学/周期表向(45 个仓库):`chemistry` 25× · `periodic-table` 11× ·
`visualization` 11× · `science` 8× · `periodic-table-of-elements` 7× · `education` 6×
- 科学教育向(30 个仓库):`education` 10× · `science` 7× · `interactive` 3× ·
`interactive-learning` 3× · `physics` 3× · `chemistry` 3× · `biology` 2× · `stem` 2×
**凭常识猜错了三个**:动手前先猜了 `chinese`、`teaching`、`static-site`,
在 75 个样本里**各出现 0 次**。这正是本 Skill 那句「用目标用户真正会搜的词,
不是你觉得贴切的词」的实例——包括写这句话的人自己也会猜错。
### 这组数据同样测不到的
它测的是**热门同类仓库打了什么 topic**,不是**打了这些就会被找到**。
和上一节同样的幸存者偏差:打了相同 topic 却没人看的仓库,搜索按星排序,根本取不到。
## 五、采集方法(可复现)
```
GET /search/repositories?q=created:>YYYY-MM-DD stars:>N&sort=stars
GET /repos/{full} → 星数、创建时间、描述、topics、license
GET /users/{owner} → 粉丝数、账号年龄、公开库数
GET /repos/{full}/readme → README 原文,取前 1200 字符作为"首屏"
GET /repos/{full}/git/trees/{branch}?recursive=1 → 文件构成
GET /search/repositories?q=<中/英文词>&sort=stars → 第四节的中英文查询对比
```
"首屏"定义为 README 前 1200 字符——对应 GitHub 上大约不滚动能看到的高度。
这是个近似值,实际取决于窗口大小和内容密度。
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 198k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |