agentleFS
lumisumlumisum/article-expert/midlife-article/SKILL.md

midlife-article

Skill2 starsChanged 3 months agoApache-2.0View on GitHub

专注 35+ 中年人生记忆与情感叙事类微信公众号文章创作技能。以用户提供的歌曲、电影或电视剧为情绪钥匙,回到真实人生现场,写出有回忆张力、因果深度与现实支点的深度好文,杜绝作品复述、流量焦虑诱导与版权化复刻。

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 1

    1. StellarLucas/article-expert★ 2100% same

    Edited · 0

    None found.

    Wrote their own version · 0

    None found.

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository8 other files it gives its agents.

The file

Files
SKILL.md

midlife-article

目标

把一段中年人的真实经历,连同用户提供的一首歌、电影或电视剧,写成一篇有人味、有回忆张力、有因果深度的公众号文章。

作品是情绪钥匙,不是被解说的对象:一段旋律或一个停住的片段,先唤起可感的声音、光线、气味、地点或关系,再让过去与此刻相撞。文章写的是“它为什么让这个人停住了”,不是歌词赏析、剧情回顾或影视评论。

研究材料只在后台校准事实。公开表达回到人物、关系、时间、责任、欲望、选择与日常生活。作者把人和因果写到位,让读者自己感觉到什么值得珍惜、承担或改变。

不可变契约

  1. 只讲透一个人生问题:具体事件只是入口,全文只回答一个中年读者真正关心的问题。
  2. 口述材料优先:用户的动作、犹豫、原话和生活细节是叙事主体,外部资料不能把它淹没。
  3. 深度来自因果:下一层必须是上一层的因;并列概念、换案例和哲学类比不计入主下潜链。
  4. 研究语言不进入正文:理论名称只在删除后会损害准确性时出现,并立刻转成生活处境。
  5. 哲学只照明,作者先判断:先由故事形成人生命题,再以东方为主、西方为辅照明;经典不替代事实,文章不停在名言上。
  6. 帮助藏在人物里:每篇只有一个主要帮助,痛点只负责认出自己。结尾不给清单、步骤、打卡或验证;读者从人物结果里自己形成判断。
  7. 只保留一个中心判断:正文不并列多个思想火花,不用固定口号和固定比喻统治全文。
  8. 配图是原创记忆插画:memory_haze_storybook_illustration,手绘平涂,禁止摄影与3D。先决定意思由人物传达还是由留白与环境传达;后者必须用全画幅远景。全文在三至六个叙事转折处留3:4占位符,只生成提示词。细节见references/stage5.md。
  9. 标题与封面为点击率服务:在正文能兑现的前提下,必须选点击欲最强的标题;封面画让人停住的那个瞬间,不画最安稳的收场。不写成影评或年龄口号。细节见references/stage6.md。
  10. 媒介与版权:默认不复写歌词或台词;作品事实只写有来源的最小范围;不凭记忆补剧情。图像不得画演员、可识别角色、片名/Logo 或标志性构图、场景、服装。
  11. 说话方式只读author.md:人称、开篇硬钩、生活化语言、反焦虑与结尾姿态以作者文件为准,其它文件不复述。
  12. 沉重之后必须回升:中前段可以下潜,最低点之后必须埋伏一处具体的向上信号——还在的人、还暖的动作、还没消失的关系或仍可向前的一步。信号来自故事里已经发生的事,不靠口号、鸡汤或布置任务。不得用灾难化、羞辱或虚构恶化逼读者行动。

流水线

完整文章依次完成 S0–S6,S7 可选。用户已给定主题或口述时从 S1 开始;要求完整公众号文章时必须走到 S6。

阶段必读产出校验
S0 选题stage0.md三个候选问题--stage 0
S1 见事stage1.mdsource_pack.json 简报、材料、证据--stage 1
S2 下潜stage2.md + author.md回声链、因果主链、Spark 草案--stage 2
S3 悟道stage3.md + author.md审计、哲学、Spark 定型--stage 3
S4 成文stage4.md + author.md无章节的 final_article.md--stage 4
S5 配图stage5.mdimage_package.json 与正文占位符--stage 5
S6 标题封面stage6.mdtitle_cover_package.json--stage 6
S7 口述语音stage7.md + author.mdvideo_package.json 与音频--stage 7

进入阶段前读取对应文件,不得凭记忆复述,也不得跳过 workflow_state.json。

S1–S6 必须逐阶段执行:

  1. 确认上一阶段状态存在且签名仍匹配;
  2. 只写当前阶段产物;
  3. 立即运行 python skills/midlife-article/scripts/validate_stage.py --stage N --topic-dir /absolute/topic/path;
  4. 只有 status=passed 才能进入下一阶段;失败则停在当前阶段修复后重跑。

workflow_state.json 由校验器独占写入。S2、S3 的新增内容写回同一个 research/source_pack.json。禁止把 S1–S6 合并为一次生成。未跑完校验时只能报告“流程未完成”。

工作目录

${REAL_USER_HOME}/wechat_articles/topics/[topic_id]/
├── research/source_pack.json
├── research/raw_pages/
├── article/final_article.md
├── assets/image_package.json
├── assets/title_cover_package.json
├── video/video_package.json
├── video/images/
├── video/audio/
└── workflow_state.json

S0 使用 ${REAL_USER_HOME}/wechat_articles/midlife_scans/[scan_id]/。

最终只交付 final_article.md,不生成 HTML。正文一个 H1,无章节标题。在线研究先发现原始页面再保存;搜索摘要不进证据池。用户口述只证明当事人这样经历过。不生成未声明的报告文件。S7 通过后,用户把对应图片放入 video/images/。

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.