agentleFS
ykjshenzhen2008-ship-itykjshenzhen2008-ship-it/commercial-real-estate-skills/skills/warm-article-daily/SKILL.md

warm-article-daily

Skill0 starsChanged 35 days agoMITView on GitHub

为「暖伴」老年人知识陪伴小程序每日生成一篇养生/生活/情感短文并入库。当被要求"更新暖伴知识库""写一篇养生文章""每日文章更新"时使用。

The file

Files
SKILL.md

暖伴 · 每日文章更新

给 60+ 岁老人看的短文,写进 $HOME/WorkBuddy/小程序/data/content.src.json。

铁律(违反即为不合格)

  1. 绝不诊断、绝不开药:不写具体药名、剂量、疗程。涉及用药一律"按大夫说的来""去社区卫生站让大夫看看"。
  2. 只讲能照做的日常小事:一段一个动作,别讲道理讲道理。
  3. 老人视角:用"您""咱们",短句,像邻家晚辈说话。不用 emoji、不用小标题、不用"首先其次"。
  4. 不改 JS 源码:内容只进 content.src.json,由脚本生成 content.generated.js。

执行流程

Step 1 · 选今天的题(跑脚本,别自己想)

cd $HOME/WorkBuddy/小程序
$HOME/.workbuddy/binaries/node/versions/20.18.0/bin/node tools/pick-topic.js

脚本会输出 {cat, sub, angle, existing, hint}:哪个子分类最缺、该写什么角度、已有标题(避免重复)。必须按它给的 cat/sub 写。

Step 2 · 写

结构固定 5–7 段,每段 40–90 字:

段位写什么
1场景切入:一件老人真会遇到的小事("早上起来嗓子干")
2–5具体怎么做,一段一个动作,能照着做
6安全兜底:拿不准就找大夫/叫孩子帮忙
7一句有温度的话收尾

字段:{ id, cat, sub, title(≤16字), summary(≤40字), readMin, content:[...] }

  • id 可省略(脚本自动生成 a-<cat>-<n>)
  • readMin 可省略(脚本按字数算)

写完存成 /tmp/warm-new-article.json。

Step 3 · 校验(先 dry-run)

$HOME/.workbuddy/binaries/node/versions/20.18.0/bin/node tools/add-article.js --file /tmp/warm-new-article.json --dry-run

脚本会挡住:医疗红线(药名/剂量/疗效承诺/诊断)、标题重复、段落过短过长、没用"您"、书面序号。 报错就改,改到全绿再进 Step 4。

Step 4 · 入库

$HOME/.workbuddy/binaries/node/versions/20.18.0/bin/node tools/add-article.js \
  --file /tmp/warm-new-article.json --quote "今日新写的一句暖心话(20-30字)"

脚本会自动:写 JSON → 重建 data/content.generated.js → 记 data/article-history.json。

Step 5 · 回归(必须做)

NODE=$HOME/.workbuddy/binaries/node/versions/20.18.0/bin/node
$NODE tools/static-check.js $HOME/WorkBuddy/小程序
$NODE tools/runtime-test-unconfigured.js

两套必须全绿。若新增的是某个子分类的第一篇文章,static-check 的"子分类无文章"告警会消失,属正常。

Step 6 · 报告

一句话:今天写了哪篇、哪个分类、现在知识库共几篇。

参考

  • 写作风格与样例:references/writing-style.md
  • 医疗合规红线:references/medical-redlines.md

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.