github-trending-writer
Skill2 starsChanged 8 months agoApache-2.0View on GitHub
Find GitHub Trending repositories, save readable repo source material locally, and draft Chinese articles in a review-first workflow. Use when the user wants to 看 GitHub Trending、整理热门开源项目、把仓库内容落成中文稿、或基于 Trending 项目做技术热点文章后再发布。
In this repository27 other files it gives its agents.
- ai-hotspot-collectorSkill
- article-multi-publisherSkill
- article-pipeline-skillSkill
- character-ip-risk-checkSkill
- chrome-extension-devSkill
- chrome-extension-e2e-automationSkill
- chrome-extension-publishSkill
- chrome-extension-social-promoSkill
- chrome-webstore-image-generatorSkill
- code-reviewSkill
- fund-querySkill
- lark-workflow-doc-to-slidesSkill
- notionSkill
- playwright-twitter-accessSkill
- post-to-xhsSkill
- rewrite-skillSkill
- small-capital-investingSkill
- technical-article-preflightSkill
- technical-article-reviewSkill
- technical-spoken-rewriteSkill
- tencent-dev-community-publisherSkill
- toutiao-publisherSkill
- url-readerSkill
- wechat-article-formatterSkill
- wechat-article-generatorSkill
- wechat-publisherSkill
- wechat-sticker-publisherSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use this skill when the user wants one repeatable workflow for:
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
GitHub Trending Writer
Use this skill when the user wants one repeatable workflow for:
- 从 GitHub Trending 选项目
- 把仓库 README / 仓库页内容落到本地候选目录
- 基于候选目录写中文技术稿
- 先 review 再修稿
- 审阅通过后再交给平台发布 skill
Unified Entry
From the repository root:
python3 github-trending-writer/scripts/run.py <command> [options]Commands:
fetch: 抓取 GitHub Trending 候选并落盘publish: 把已审阅候选交给 article-multi-publisher
Workflow
1. 抓取候选
默认抓当天 Trending:
python3 github-trending-writer/scripts/run.py fetch按语言或周期抓取:
python3 github-trending-writer/scripts/run.py fetch \
--language python \
--since weekly \
--limit 4底层脚本也可直接调用:
python3 github-trending-writer/scripts/fetch_github_trending.py抓取阶段会:
- 抓取 GitHub Trending 页面
- 解析仓库名、描述、语言、总 stars、forks、今日新增
- 使用 repo-local url-reader 读取 README / 仓库页内容
- 过滤错误页、空白页或无效内容
- 为每个候选生成:
content.mdraw.mdmeta.json
- 在输出根目录生成:
manifest.jsonREADME.md
默认输出目录:
tmp/github-trending-review/{YYYY-MM-DD}/curated/2. 向用户展示候选
优先展示:
- 编号
- 仓库名
- Trending 周期
- 语言、总 stars、今日新增
- 仓库链接
- 本地
content.md路径
如果 README 明显无内容、错误页或只有占位文本,不要推荐给用户。
3. 读取选中的候选内容
用户选定某篇后:
- 先读对应目录里的
content.md - 再读 references/style-guide.md
- 根据用户意图决定稿型:
- 默认:
解读型 - 用户明确说“更想看原文”“先翻译”时:
翻译整理型
- 默认:
4. 写稿后先 Review,再给用户看修订稿
默认不要把首稿当成可发稿。
标准顺序:
- 先把主稿写到候选目录
- 立即调用 technical-article-review
- 除非用户明确要求
review-only、只审不改、先给意见别动稿,否则直接按审阅意见修改源稿 - 只把 review 后的修订稿发给用户审阅
推荐文件名:
article.mdtranslation-draft.zh.mdreview-draft.zh.md
5. 审阅通过后再发布
python3 github-trending-writer/scripts/run.py publish <candidate-dir>常用参数:
python3 github-trending-writer/scripts/run.py publish <candidate-dir> \
--platforms wechat,xhs,toutiao \
--dry-run如果输入的是候选根目录,也可以按编号发:
python3 github-trending-writer/scripts/run.py publish <curated-root> \
--index 1 \
--source article.md发布边界:
github-trending-writer只负责候选发现、落盘、写稿约定和成稿选择- 实际平台发布委托给 article-multi-publisher
- 微信平台的作者、封面、API/browser 逻辑继续归 wechat-publisher
Writing Rules
解读型
适用于:
- 用户想知道“这个仓库为什么上 Trending”
- 更在意定位、价值和使用场景
要求:
- 先说“这个仓库在解决什么问题”
- 再讲“为什么最近火”
- 再讲“适合谁”和“边界在哪里”
- 文末保留仓库链接或仓库名
翻译整理型
适用于:
- 用户明确想看更完整的原始信息
- README 本身结构很好
要求:
- 先用中文讲清仓库在做什么
- 主体按 README 或目录结构整理
- 结尾只补少量解释,不要喧宾夺主
Quality Bar
- 不要把 README 很弱的仓库硬写成深度项目分析
- 不要把 Trending 上榜误写成官方背书
- 不要虚构仓库没有的能力
- 区分 README 原始信息和作者解读
- 不要把未 review 的首稿推进到 HTML / 封面 / 发布
References
- 风格说明:references/style-guide.md
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.

