agentleFS
ythx-101ythx-101/paper-reader/llms.txt

paper-reader

llms.txt21 starsChanged 3 days agoMITView on GitHub

给 Claude Code 等 AI agent 用的论文解读技能(skill):只走合法途径获取原文,整理每个数字都有出处的事实表,按 PaperBanana 五角色流程出图,强模型评审忠实度,再用 Kami 排版成 X / 小红书图解组图。 An agent skill that turns an academic paper into a faithful, sourced illustrated carousel for X or Xiaohongshu.

  • In this repositoryOne other file it gives its agents.

The file

Files
llms.txt
# paper-reader

> 给 Claude Code 等 AI agent 用的论文解读技能(skill):只走合法途径获取原文,整理每个数字都有出处的事实表,按 PaperBanana 五角色流程出图,强模型评审忠实度,再用 Kami 排版成 X / 小红书图解组图。
> An agent skill that turns an academic paper into a faithful, sourced illustrated carousel for X or Xiaohongshu.

## What it does

1. 合法获取原文(Unpaywall / PMC / arXiv → 作者主页 → 免费补充材料 → 开放获取评论与综述 → 学位论文 → 起草索取单行本的邮件),过程记进 `fulltext/paths.md`。
2. 整理带出处的事实表 `facts.md`:结论 | 数字 | 出自哪张图或表 | 来源 URL。
3. 按 PaperBanana 五角色出图:检索 → 规划(`plan.md`)→ 风格 → 出图 → 评审(`reviews/`)。
4. 忠实度 P0 闸门:编数字、箭头画反、说法超出原文都是 P0,不清零不交付。
5. 用 Kami 排版输出 PNG:X 发 4 张,小红书一般 6–9 张。

## Install

- curl:`curl -L -o ~/.claude/skills/paper-reader/SKILL.md https://raw.githubusercontent.com/ythx-101/paper-reader/main/SKILL.md`(先 `mkdir -p ~/.claude/skills/paper-reader`)
- git clone:`git clone https://github.com/ythx-101/paper-reader ~/.claude/skills/paper-reader`
- Kami(出图依赖):`npx skills add tw93/kami -a claude-code codex cursor -g -y`

## Key rules

- 只走合法途径拿原文,不使用 Sci-Hub、LibGen、Anna's Archive 等盗版站。
- 新闻稿只当线索,上图数字必须回原文或补充材料核对。
- 事实表每个数字都要有出处,没有出处的说法不上图。
- 数据图一律用代码画(matplotlib 或 SVG)。
- 忠实度 P0 没清零不交付。
- 类比与引申要明确标注「我的引申,不是论文结论」。
- 不替用户发帖,成品交给用户自己发。

## Links

- [README](https://github.com/ythx-101/paper-reader/blob/main/README.md)
- [SKILL.md](https://raw.githubusercontent.com/ythx-101/paper-reader/main/SKILL.md)
- [Kami](https://github.com/tw93/Kami)
- [PaperBanana (arXiv 2601.23265)](https://arxiv.org/abs/2601.23265)

## License

MIT

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.