Yuan1z0825/nature-skillsPublic

符合nature论文学术表达和科研绘图的Skill

AI summary: A collection of AI skills for parsing, analyzing, and synthesizing scientific literature from Nature.

Stars
33.8K
+361 today
Forks
1.9K
Watchers
39
Open issues
0
Open PRs
2
Contributors
~26
Commits
724
Branches
1

PythonApache-2.0Created Apr 24, 2026Last push today+1.3K stars this week+2K this month

Star history

since Apr 26, 2026
010K20K30KApr 2026May 2026Jul 2026Aug 2026
33.8K stars as of Aug 7, 2026, tracked back to Apr 26, 2026. Historical curve reconstructed from public GitHub event archives, calibrated to the current total.

Contribution activity

commits per day, last 52 weeks
AugSepOctNovDecJanFebMarAprMayJunJulMonWedFri2025-08-03: 0 commits2025-08-04: 0 commits2025-08-05: 0 commits2025-08-06: 0 commits2025-08-07: 0 commits2025-08-08: 0 commits2025-08-09: 0 commits2025-08-10: 0 commits2025-08-11: 0 commits2025-08-12: 0 commits2025-08-13: 0 commits2025-08-14: 0 commits2025-08-15: 0 commits2025-08-16: 0 commits2025-08-17: 0 commits2025-08-18: 0 commits2025-08-19: 0 commits2025-08-20: 0 commits2025-08-21: 0 commits2025-08-22: 0 commits2025-08-23: 0 commits2025-08-24: 0 commits2025-08-25: 0 commits2025-08-26: 0 commits2025-08-27: 0 commits2025-08-28: 0 commits2025-08-29: 0 commits2025-08-30: 0 commits2025-08-31: 0 commits2025-09-01: 0 commits2025-09-02: 0 commits2025-09-03: 0 commits2025-09-04: 0 commits2025-09-05: 0 commits2025-09-06: 0 commits2025-09-07: 0 commits2025-09-08: 0 commits2025-09-09: 0 commits2025-09-10: 0 commits2025-09-11: 0 commits2025-09-12: 0 commits2025-09-13: 0 commits2025-09-14: 0 commits2025-09-15: 0 commits2025-09-16: 0 commits2025-09-17: 0 commits2025-09-18: 0 commits2025-09-19: 0 commits2025-09-20: 0 commits2025-09-21: 0 commits2025-09-22: 0 commits2025-09-23: 0 commits2025-09-24: 0 commits2025-09-25: 0 commits2025-09-26: 0 commits2025-09-27: 0 commits2025-09-28: 0 commits2025-09-29: 0 commits2025-09-30: 0 commits2025-10-01: 0 commits2025-10-02: 0 commits2025-10-03: 0 commits2025-10-04: 0 commits2025-10-05: 0 commits2025-10-06: 0 commits2025-10-07: 0 commits2025-10-08: 0 commits2025-10-09: 0 commits2025-10-10: 0 commits2025-10-11: 0 commits2025-10-12: 0 commits2025-10-13: 0 commits2025-10-14: 0 commits2025-10-15: 0 commits2025-10-16: 0 commits2025-10-17: 0 commits2025-10-18: 0 commits2025-10-19: 0 commits2025-10-20: 0 commits2025-10-21: 0 commits2025-10-22: 0 commits2025-10-23: 0 commits2025-10-24: 0 commits2025-10-25: 0 commits2025-10-26: 0 commits2025-10-27: 0 commits2025-10-28: 0 commits2025-10-29: 0 commits2025-10-30: 0 commits2025-10-31: 0 commits2025-11-01: 0 commits2025-11-02: 0 commits2025-11-03: 0 commits2025-11-04: 0 commits2025-11-05: 0 commits2025-11-06: 0 commits2025-11-07: 0 commits2025-11-08: 0 commits2025-11-09: 0 commits2025-11-10: 0 commits2025-11-11: 0 commits2025-11-12: 0 commits2025-11-13: 0 commits2025-11-14: 0 commits2025-11-15: 0 commits2025-11-16: 0 commits2025-11-17: 0 commits2025-11-18: 0 commits2025-11-19: 0 commits2025-11-20: 0 commits2025-11-21: 0 commits2025-11-22: 0 commits2025-11-23: 0 commits2025-11-24: 0 commits2025-11-25: 0 commits2025-11-26: 0 commits2025-11-27: 0 commits2025-11-28: 0 commits2025-11-29: 0 commits2025-11-30: 0 commits2025-12-01: 0 commits2025-12-02: 0 commits2025-12-03: 0 commits2025-12-04: 0 commits2025-12-05: 0 commits2025-12-06: 0 commits2025-12-07: 0 commits2025-12-08: 0 commits2025-12-09: 0 commits2025-12-10: 0 commits2025-12-11: 0 commits2025-12-12: 0 commits2025-12-13: 0 commits2025-12-14: 0 commits2025-12-15: 0 commits2025-12-16: 0 commits2025-12-17: 0 commits2025-12-18: 0 commits2025-12-19: 0 commits2025-12-20: 0 commits2025-12-21: 0 commits2025-12-22: 0 commits2025-12-23: 0 commits2025-12-24: 0 commits2025-12-25: 0 commits2025-12-26: 0 commits2025-12-27: 0 commits2025-12-28: 0 commits2025-12-29: 0 commits2025-12-30: 0 commits2025-12-31: 0 commits2026-01-01: 0 commits2026-01-02: 0 commits2026-01-03: 0 commits2026-01-04: 0 commits2026-01-05: 0 commits2026-01-06: 0 commits2026-01-07: 0 commits2026-01-08: 0 commits2026-01-09: 0 commits2026-01-10: 0 commits2026-01-11: 0 commits2026-01-12: 0 commits2026-01-13: 0 commits2026-01-14: 0 commits2026-01-15: 0 commits2026-01-16: 0 commits2026-01-17: 0 commits2026-01-18: 0 commits2026-01-19: 0 commits2026-01-20: 0 commits2026-01-21: 0 commits2026-01-22: 0 commits2026-01-23: 0 commits2026-01-24: 0 commits2026-01-25: 0 commits2026-01-26: 0 commits2026-01-27: 0 commits2026-01-28: 0 commits2026-01-29: 0 commits2026-01-30: 0 commits2026-01-31: 0 commits2026-02-01: 0 commits2026-02-02: 0 commits2026-02-03: 0 commits2026-02-04: 0 commits2026-02-05: 0 commits2026-02-06: 0 commits2026-02-07: 0 commits2026-02-08: 0 commits2026-02-09: 0 commits2026-02-10: 0 commits2026-02-11: 0 commits2026-02-12: 0 commits2026-02-13: 0 commits2026-02-14: 0 commits2026-02-15: 0 commits2026-02-16: 0 commits2026-02-17: 0 commits2026-02-18: 0 commits2026-02-19: 0 commits2026-02-20: 0 commits2026-02-21: 0 commits2026-02-22: 0 commits2026-02-23: 0 commits2026-02-24: 0 commits2026-02-25: 0 commits2026-02-26: 0 commits2026-02-27: 0 commits2026-02-28: 0 commits2026-03-01: 0 commits2026-03-02: 0 commits2026-03-03: 0 commits2026-03-04: 0 commits2026-03-05: 0 commits2026-03-06: 0 commits2026-03-07: 0 commits2026-03-08: 0 commits2026-03-09: 0 commits2026-03-10: 0 commits2026-03-11: 0 commits2026-03-12: 0 commits2026-03-13: 0 commits2026-03-14: 0 commits2026-03-15: 0 commits2026-03-16: 0 commits2026-03-17: 0 commits2026-03-18: 0 commits2026-03-19: 0 commits2026-03-20: 0 commits2026-03-21: 0 commits2026-03-22: 0 commits2026-03-23: 0 commits2026-03-24: 0 commits2026-03-25: 0 commits2026-03-26: 0 commits2026-03-27: 0 commits2026-03-28: 0 commits2026-03-29: 0 commits2026-03-30: 0 commits2026-03-31: 0 commits2026-04-01: 0 commits2026-04-02: 0 commits2026-04-03: 0 commits2026-04-04: 0 commits2026-04-05: 0 commits2026-04-06: 0 commits2026-04-07: 0 commits2026-04-08: 0 commits2026-04-09: 0 commits2026-04-10: 0 commits2026-04-11: 0 commits2026-04-12: 0 commits2026-04-13: 0 commits2026-04-14: 0 commits2026-04-15: 0 commits2026-04-16: 0 commits2026-04-17: 0 commits2026-04-18: 0 commits2026-04-19: 0 commits2026-04-20: 0 commits2026-04-21: 0 commits2026-04-22: 0 commits2026-04-23: 0 commits2026-04-24: 9 commits2026-04-25: 0 commits2026-04-26: 0 commits2026-04-27: 0 commits2026-04-28: 0 commits2026-04-29: 9 commits2026-04-30: 1 commit2026-05-01: 9 commits2026-05-02: 4 commits2026-05-03: 2 commits2026-05-04: 2 commits2026-05-05: 0 commits2026-05-06: 14 commits2026-05-07: 16 commits2026-05-08: 16 commits2026-05-09: 16 commits2026-05-10: 22 commits2026-05-11: 11 commits2026-05-12: 14 commits2026-05-13: 12 commits2026-05-14: 17 commits2026-05-15: 10 commits2026-05-16: 17 commits2026-05-17: 13 commits2026-05-18: 9 commits2026-05-19: 7 commits2026-05-20: 13 commits2026-05-21: 9 commits2026-05-22: 8 commits2026-05-23: 12 commits2026-05-24: 13 commits2026-05-25: 10 commits2026-05-26: 12 commits2026-05-27: 11 commits2026-05-28: 9 commits2026-05-29: 24 commits2026-05-30: 21 commits2026-05-31: 22 commits2026-06-01: 7 commits2026-06-02: 3 commits2026-06-03: 0 commits2026-06-04: 0 commits2026-06-05: 0 commits2026-06-06: 0 commits2026-06-07: 3 commits2026-06-08: 0 commits2026-06-09: 7 commits2026-06-10: 1 commit2026-06-11: 0 commits2026-06-12: 0 commits2026-06-13: 0 commits2026-06-14: 8 commits2026-06-15: 0 commits2026-06-16: 10 commits2026-06-17: 0 commits2026-06-18: 12 commits2026-06-19: 14 commits2026-06-20: 2 commits2026-06-21: 0 commits2026-06-22: 1 commit2026-06-23: 0 commits2026-06-24: 0 commits2026-06-25: 0 commits2026-06-26: 1 commit2026-06-27: 0 commits2026-06-28: 1 commit2026-06-29: 6 commits2026-06-30: 5 commits2026-07-01: 4 commits2026-07-02: 1 commit2026-07-03: 2 commits2026-07-04: 2 commits2026-07-05: 8 commits2026-07-06: 3 commits2026-07-07: 1 commit2026-07-08: 3 commits2026-07-09: 2 commits2026-07-10: 1 commit2026-07-11: 0 commits2026-07-12: 7 commits2026-07-13: 6 commits2026-07-14: 1 commit2026-07-15: 6 commits2026-07-16: 21 commits2026-07-17: 13 commits2026-07-18: 22 commits2026-07-19: 7 commits2026-07-20: 2 commits2026-07-21: 1 commit2026-07-22: 6 commits2026-07-23: 0 commits2026-07-24: 0 commits2026-07-25: 2 commits2026-07-26: 0 commits2026-07-27: 1 commit2026-07-28: 3 commits2026-07-29: 0 commits2026-07-30: 0 commits2026-07-31: 0 commits2026-08-01: 0 commits
600 commits in the last yearLessMore

Signals and awards

derived from tracked data
  • Widely adopted

    33,823 stars

  • Breakout launch

    33,823 stars in 106 days

  • Very active

    600 commits in 52 weeks

  • Outside contributions

    73% of recent commits from the community

  • Permissive license

    Apache-2.0

  • Continuous integration

    Automated checks passing

  • Repeat trending

    3 trending appearances

What nature-skills does

Nature-skills is a specialized repository containing a suite of AI skills, prompts, and tools designed specifically for interacting with scientific literature, particularly articles from Nature and similar high-impact journals. It provides structured approaches for large language models to accurately extract data, summarize complex methodologies, and synthesize findings from dense academic texts. The project includes specialized parsers for scientific PDFs, custom system prompts tuned for academic rigor, and workflows for conducting automated literature reviews. It aims to reduce AI hallucinations when dealing with highly technical scientific information by enforcing strict grounding techniques.

This repository is primarily for academic researchers, data scientists, and students working with scientific literature. It is useful for anyone leveraging AI to accelerate their academic research and literature review processes.

  • Academic prompts: Features highly tuned system prompts designed to enforce academic rigor and reduce hallucinations in LLMs.
  • Methodology extraction: Includes specific skills for identifying and summarizing complex experimental procedures from papers.
  • Data synthesis tools: Provides workflows for aggregating and synthesizing data points across multiple scientific articles.
  • PDF parsing utilities: Integrates specialized tools for accurately extracting text, tables, and figures from academic PDFs.
  • Citation management: Ensures that generated summaries properly attribute information to specific sections of the source text.
  • Domain-specific formatting: Outputs summaries in standard academic formats suitable for research notes or literature reviews.

Where teams use it

Automated literature reviews

Researchers use the skills to rapidly scan and summarize dozens of papers to identify trends in their specific field.

Methodology analysis

Scientists employ the tools to extract and compare experimental procedures across different studies.

Academic reading assistant

Students use the tailored prompts to help them break down and understand dense, highly technical scientific articles.

Data extraction pipeline

Data scientists use the parsing utilities to programmatically extract datasets and tables from published research.

Getting started: Clone the repository and import the specific prompt templates into your preferred AI agent framework.

README

main branch

nature-skills 面向全球 AI 学者收录可复用科研技能,强调真实问题解决、可验证工作流与可直接使用的科研产物。

目录

1. 项目发起人与运营信息

1.1 创始人介绍

大家好,我是 nature-skills 的创立者袁一哲。感谢大家持续关注本项目。我们在抖音更新了许多视频教程,大家可以根据名称检索查看,希望能够真正帮助到科研工作。

1.2 知识星球

知识星球名称:Nature Skills 以及背后的哲学

Nature Skills 知识星球

1.3 自营 GPT / Claude 代充与成品号

严格筛选渠道商,提供完全正规的充值渠道与服务。欢迎访问 Nature AI 充值卡网(已上线plus一年代充)

Nature AI GPT 与 Claude 代充及成品号服务
Nature AI 充值卡网
https://apiciyuan.top/
250d280342f34902a527721a118ac52e 微信客服扫码添加

1.4 商务合作

如有商务合作意向,欢迎发送邮件至 natureskills2026@outlook.com

2. Skills 主要开发者

开发者 项目角色 主要方向与贡献 主页与联系
袁一哲 创始人 / 维护者 项目发起、技能体系设计与社区运营
马昕瑞 核心开发者 Skills 日常维护 Gmail
胡彬 主要贡献者 Agentic Agent Email

3. 项目理念与社区

3.1 自己的一些浅薄观点

  • 最近发现,我设计的 Nature Skills 被谷歌 DeepMind 关注并借鉴,他们参考了其中的引用体系、脚本思路以及技能设计哲学,推出了 Science Skills。说实话,这让我挺欣慰的——当国外的顶尖 AI 机构开始从我们的工作中汲取灵感时,说明中国开发者的原创思想正在被世界看见。这不是被复制的失落,而是中国力量在开源土壤里生根后,自然向外生长出的影响力。
  • 我们设计 Skills 的重心,从来不是要求每个人都来啃透这套思想,而是这套思想本身就具备被机器理解并复用的能力。你如果想创立一个全新的 Skill,或者把它适配到自己的专属领域,根本不需要从头学起——直接把 Nature Skills 的 GitHub 地址发给 Codex,它就能自动学习其中的设计精髓,帮你完成新 Skill 的创建和修改。这才是思想的真正解放:它不再依赖口口相传,而是通过 AI 直接流淌进每一个需要它的角落。
  • Nature Skills 真正的价值,或许并不在于那些具体的技能模块,而在于它悄悄推开了一扇新的大门——它让很多人第一次意识到,原来可以借助 Codex 或智能体来操控本地电脑做科研。我有幸见证并陪伴了许多人完成科研范式的转变。当他们惊叹“原来科研还可以这样去做”的那一刻,这种认知上的破壁和解放,远比 Skills 本身更让我觉得有意义。它不是一个工具的成功,而是一种新的思考方式开始在人群中蔓延。
  • 在当下,几乎所有实用的工具都可以被提炼为标准化流程,而标准化流程恰好可以封装成可复用技能。

3.2 视频教程与社区

视频教程请关注抖音
抖音视频教程
Agent 科研交流群
Agent 科研交流群
袁博个人微信
袁博个人微信

4. 快速开始

安装完成后,可以直接把论文、段落、审稿意见或任务描述交给 Agent。下面这些提示词可以直接复制使用:

想做什么 直接这样说
读论文 / 中英文对照 把这篇 PDF 做成图文对应的中英文对照 Markdown reader。
生成文献汇报 PPT 把这篇论文做成中文组会汇报 PPT,保留关键图件和来源标注。
润色或翻译论文段落 把这段中文改写成 Nature 风格英文,保持学术含义不变。
写摘要、引言或讨论 根据这些结果和图件,帮我起草 Nature 风格的摘要和引言。
预投稿审稿模拟 从 Nature 审稿人视角评估这篇稿件,给出三份 reviewer reports。
回复审稿意见 根据这封返修邮件,帮我写逐点回复、cover letter,并标出修改稿需要标红的位置。
查文献、他引和引用者画像 整理这篇文章的引用数、严格他引数、DOI,并看引用者里有没有院士、Fellow 或领域大牛。
做科研图或论文示意图 根据这段方法和结果,帮我生成投稿级科研图或论文示意图草稿。

如果你不确定该用哪个技能,直接描述任务即可;如果已经知道技能名,可以在提示词中明确写“使用 nature-reader”或“使用 nature-response”。

5. 安装

nature-skills 是一组围绕 SKILL.md 组织的可复用技能包。skills/ 下的每个顶层技能目录都是一个可安装单元,例如 nature-*nature-shared 是供其他技能读取的共享支持包,默认不作为独立触发技能计入技能索引。

5.1 npx skills 安装方式

需要先安装 Node.js 18 或更高版本。无需全局安装 CLI;先查看仓库中可安装的技能名:

npx skills add Yuan1z0825/nature-skills --list

把全部技能全局安装到 Codex。nature-shared 会随全量安装一起加入,因此依赖共享参考资料的技能也能正常工作:

npx skills add Yuan1z0825/nature-skills --global --agent codex --skill '*' --yes --copy

只为当前项目安装一个独立技能时,省略 --global。例如:

npx skills add Yuan1z0825/nature-skills --agent codex --skill nature-figure --yes --copy

单独安装 nature-readernature-paper2pptnature-polishingnature-writing 时,同时选择共享支持包:

npx skills add Yuan1z0825/nature-skills --global --agent codex \
  --skill nature-reader --skill nature-shared --yes --copy

也可以把全部技能安装到 CLI 支持的所有 agent:

npx skills add Yuan1z0825/nature-skills --all

检查全局安装结果并更新:

npx skills list --global --agent codex --json
npx skills update --global --yes

只更新一个技能,或只更新当前项目中的技能:

npx skills update nature-reader --global --yes
npx skills update --project --yes

技能选择参数使用 --list 显示的 frontmatter 名称;例如目录 nature-proposal-writer 当前显示为 researchwritenpx skills 管理的是技能文件,Python、R、浏览器或 MCP 等可选运行依赖仍需按下文说明单独配置。

5.2 Claude Code 安装方式

Claude Code 不能直接使用 scripts/update-codex-skills.sh,因为这个脚本只负责同步到 Codex 的 ~/.codex/skills/。用于 Claude Code 时,推荐保留一个稳定的本地 clone,再用 subagent 或 slash command 指向真实的 skills/*/SKILL.md。这样不会破坏技能目录结构,也能继续读取 references/static/manifest.yaml、脚本、资产和 skills/nature-shared/

如果还没有安装 Claude Code:

npm install -g @anthropic-ai/claude-code
claude

先把仓库 clone 到一个稳定路径:

mkdir -p ~/ai-skills
cd ~/ai-skills
git clone https://github.com/Yuan1z0825/nature-skills.git

推荐方式:为常用技能创建 Claude Code subagent wrapper。以 nature-reader 为例:

mkdir -p ~/.claude/agents
cat > ~/.claude/agents/nature-reader.md <<'EOF'
---
name: nature-reader
description: Use for Chinese-English paper reading, figure-aware translation, and source-grounded paper notes.
---

When invoked, first read `~/ai-skills/nature-skills/skills/nature-reader/SKILL.md` and follow it as the governing workflow.
Read supporting files from `~/ai-skills/nature-skills/skills/nature-reader/` and `~/ai-skills/nature-skills/skills/nature-shared/` only when needed.
Do not replace this skill with a generic paper-reading response.
EOF

然后开启新的 Claude Code 会话,直接请求使用这个 subagent:

Use the nature-reader subagent to turn this paper into a Chinese-English Markdown reader.

如果你更喜欢 slash command,也可以创建命令 wrapper:

mkdir -p ~/.claude/commands
cat > ~/.claude/commands/nature-reader.md <<'EOF'
Read `~/ai-skills/nature-skills/skills/nature-reader/SKILL.md` first and follow it strictly.
Read directly needed supporting files from `~/ai-skills/nature-skills/skills/nature-reader/` and `~/ai-skills/nature-skills/skills/nature-shared/`.

$ARGUMENTS
EOF

在 Claude Code 中使用:

/nature-reader 把这篇论文做成中英文对照的完整 Markdown reader。

安装其他技能时,把示例中的 nature-reader 换成对应目录名即可,例如 nature-polishingnature-writingnature-reviewernature-responsenature-figure。后续更新只需要:

cd ~/ai-skills/nature-skills
git pull

只要 wrapper 仍然指向这个稳定 clone 路径,就不需要重复复制技能文件。

自动更新(可选)

如果你希望 Claude Code 每次开启会话时自动拉取上游更新,可以用 scripts/autoupdate-skills.sh 配合一个 SessionStart 钩子。

这套方式把技能直接复制~/.claude/skills/(Claude Code 会自动发现该目录,技能以目录名直接加载),而不是使用上面的 wrapper。两种方式二选一即可。

先保留一个专用的稳定 clone(只用于同步技能,请不要在里面做开发提交):

mkdir -p ~/ai-skills
git clone https://github.com/Yuan1z0825/nature-skills.git ~/ai-skills/nature-skills

首次安装,把技能复制进 Claude Code 的技能目录:

~/ai-skills/nature-skills/scripts/autoupdate-skills.sh --force

然后在 ~/.claude/settings.json 里加一个 SessionStart 钩子(若已有 hooks,把这一项合并进去,不要整体覆盖):

{
  "hooks": {
    "SessionStart": [
      {
        "hooks": [
          {
            "type": "command",
            "command": "$HOME/ai-skills/nature-skills/scripts/autoupdate-skills.sh",
            "async": true,
            "timeout": 120
          }
        ]
      }
    ]
  }
}

async: true 让它在后台运行、不阻塞启动。脚本自带保护:默认 6 小时内不重复联网检查、断网或拉取失败自动跳过(exit 0,绝不卡住会话)、只有上游 HEAD 真正变化时才重新同步、并且拒绝在有未提交改动的 clone 上强行前进。拉到的新版会在下一次开启会话时生效(当前会话的技能已经加载完毕)。运行日志在 ~/.local/state/nature-skills/autoupdate.log

目标目录与检查频率都可配置:

# 默认同步到 ~/.claude/skills;用 --dest 指到别处,例如 Codex:
~/ai-skills/nature-skills/scripts/autoupdate-skills.sh --dest ~/.codex/skills
# 只在最多每小时检查一次:
~/ai-skills/nature-skills/scripts/autoupdate-skills.sh --throttle 3600

5.3 Codex 安装方式

推荐使用仓库自带脚本安装或更新 Codex skills。脚本会同步 skills/ 下所有顶层技能目录,并在复制后做 diff 验证;它不会覆盖其他无关 Codex skills。

git clone https://github.com/Yuan1z0825/nature-skills.git
cd nature-skills
scripts/update-codex-skills.sh --pull

如果已经 clone 过仓库:

cd nature-skills
scripts/update-codex-skills.sh --pull

验证当前 Codex 安装是否和这个 checkout 一致:

scripts/update-codex-skills.sh --check

如果你长期用这个脚本更新,并希望清理上游已经删除的旧技能目录:

scripts/update-codex-skills.sh --pull --prune

--prune 只会删除以前由这个脚本记录过、但当前仓库已经不再包含的目录。第一次运行没有历史记录时,它不会猜测删除旧目录。

也可以把仓库链接交给 Codex,让 Codex 执行安装脚本。推荐提示词:

请从这个仓库安装 Codex skills:
https://github.com/Yuan1z0825/nature-skills.git

请 clone 仓库后运行 scripts/update-codex-skills.sh --pull。
安装后再运行 scripts/update-codex-skills.sh --check 验证。
请保留 skills/ 下的完整技能目录,不要只复制 SKILL.md。

如果只安装单个技能,请明确说明技能名:

只安装这个仓库里的 nature-reader:
https://github.com/Yuan1z0825/nature-skills.git

如果该技能需要共享文件,也请一并安装 skills/nature-shared。

关键规则:保留完整目录结构。请复制或引用整个技能文件夹,而不是只复制 SKILL.md,因为许多技能依赖 references/static/manifest.yaml、脚本、资产或共享文件。

安装脚本不会自动安装 Python 依赖。需要使用相关脚本或 MCP 服务时,再按需安装:

python -m pip install -r skills/nature-paper-to-patent/requirements.txt
python -m pip install -r skills/nature-paper-to-patent/scripts/disclosure/requirements-cnipa.txt  # 可选:国知局公布公告检索
python -m pip install -r skills/nature-academic-search/mcp-server/requirements.txt

如果启用 nature-paper-to-patent 的国知局公布公告检索,还需要执行 python -m playwright install chromium

nature-academic-search 的 MCP 服务还需要单独配置 PUBMED_EMAIL,Scopus / ScienceDirect 等可选 provider 需要使用本机凭据配置,不要把 API key 写入仓库文件。

安装后,请开启一个新的 Codex 会话,然后自然描述任务,例如:

把这篇论文做成中英文对照的完整 Markdown reader。
把这篇论文做成中文PPT。

如果你使用 OpenClaw、OpenCode、Hermes 等开源 agent / 编程框架,请看 OpenClaw / OpenCode / Hermes 接入教程

自动更新(可选)

Codex 支持全局 SessionStart hook。保留一个专用 clone 后,可以在每次启动或恢复 Codex 会话时检查更新,并把新版同步到 ~/.codex/skills/

先创建专用 clone 并完成首次同步:

mkdir -p ~/.codex
git clone https://github.com/Yuan1z0825/nature-skills.git ~/.codex/.nature-skills-src
~/.codex/.nature-skills-src/scripts/autoupdate-skills.sh \
  --dest ~/.codex/skills --force

然后创建或合并 ~/.codex/hooks.json

{
  "hooks": {
    "SessionStart": [
      {
        "matcher": "startup|resume",
        "hooks": [
          {
            "type": "command",
            "command": "/bin/bash \"$HOME/.codex/.nature-skills-src/scripts/autoupdate-skills.sh\" --dest \"$HOME/.codex/skills\"",
            "timeout": 75,
            "statusMessage": "Checking Nature Skills updates"
          }
        ]
      }
    ]
  }
}

hooks.json 中已有其他 hook,请合并 SessionStart 项,不要整体覆盖。首次启用或修改 hook 后,在 Codex 中运行 /hooks 检查并信任它。Codex 当前按同步方式执行 command hook,因此这里依靠脚本自带的 6 小时节流、60 秒网络保护和断网自动跳过,避免每次会话都重复联网或因更新失败阻断启动。

更新日志位于 ~/.local/state/nature-skills/autoupdate.log。拉取到的新技能通常在下一次会话中完整生效。

5.4 其他 Agent 场景

OpenClaw、OpenCode、Hermes 的具体接入方式见 OpenClaw / OpenCode / Hermes 接入教程

用于其他 agent 时,建议保留一个稳定的仓库 clone,再创建轻量 subagent、slash command 或 custom prompt wrapper,指向真实的 skills/*/SKILL.md,并保留 skills/nature-shared/

手动或其他 agent 使用时:

  1. 将完整技能目录复制到你的 prompt library 或项目中。
  2. 保留 SKILL.mdmanifest.yamlstatic/references/、脚本、资产和需要的 skills/nature-shared/ 文件。
  3. 如目标 agent 有自己的格式要求,可调整 frontmatter 和正文结构。

6. 技能索引

当前 skills/ 下包含以下可触发技能;skills/nature-shared/ 是共享内容目录,不计入技能索引。点击技能名或“详情页”可以进入每个 skill 的单独说明页面。

技能 状态 用途 触发词 详情页
nature-figure Stable 面向 Nature / 高影响力期刊的 Python 或 R 投稿级科研图工作流,内置 figures4papers demo,并支持通过 OpenRouter GPT Image 2 生成论文示意图草稿 “Nature figure”, “投稿级图片”, “publication plot”, “scientific figure”, “figures4papers”, “论文示意图”, “GPT Image 2” 详情
nature-polishing Stable 将学术文本润色、重构或翻译为 Nature 风格英文 “Nature style”, “润色”, “academic writing”, “论文英文” 详情
nature-writing Draft 起草 Nature 风格手稿章节,并重建论文论证 “Nature writing”, “写摘要”, “写引言”, “manuscript draft”, “论文写作” 详情
nature-reviewer Draft 从审稿人视角模拟 Nature 风格评审,输出三份 reviewer reports 和综合意见 “Nature reviewer”, “预投稿评审”, “reviewer report”, “审稿人视角评估” 详情
nature-citation Beta 检索严格限定在 Nature / CNS 系列的支撑文献,并导出 ENW、RIS 或 Zotero RDF “Nature citation”, “CNS citation”, “分段引用”, “支撑文献”, “Zotero RDF” 详情
nature-data Draft 准备 Data Availability statement、数据仓储方案和 FAIR 检查 “Data Availability”, “数据可用性”, “repository”, “FAIR metadata” 详情
nature-statistics Draft 审查、改写或起草 Nature / 高影响力期刊投稿中的统计报告,覆盖样本量、独立分析单位、重复数、p 值、多重比较、效应量、置信区间、图注统计和审稿人统计意见 “Nature statistics”, “统计审查”, “statistical analysis”, “p value”, “sample size”, “replicates”, “multiple comparisons”, “图注统计”, “统计分析小节” 详情
nature-reader Beta 生成带来源锚点、图文对应和中英文对照的全文 Markdown reader “nature reader”, “全文 Markdown”, “原文对照”, “图文对应”, “全文翻译” 详情
nature-paper-card Beta 精读单篇论文并生成有来源约束的 01–16 节 Paper Card,覆盖方法逻辑、实验—结论证据链、结论边界、批判性分析和可检验研究想法 “nature paper card”, “论文精读”, “Paper Card”, “证据链”, “结论边界” 详情
nature-response Beta 解析返修邮件,起草、审查和修改返修 cover letter、逐点回复审稿人的 response letter、标红修改稿,并提供 LaTeX 模板 “response to reviewers”, “rebuttal letter”, “cover letter”, “major revision”, “返修邮件”, “审稿意见回复”, “修回信”, “LaTeX 模板” 详情
nature-paper2ppt Beta 从科研论文生成中文 PPTX 文献汇报 deck “paper PPT”, “journal club”, “paper to slides”, “论文汇报” 详情
nature-paper-to-patent Beta 从论文、技术报告或项目材料生成有证据约束的中国发明专利草稿,并支持专利点挖掘、查新和技术交底书迭代 “paper to patent”, “Chinese patent”, “论文转专利”, “权利要求书”, “技术交底书”, “专利点” 详情
nature-ref-verifier Stable 参考文献多源交叉验证:逐字段对比作者/标题/年份/卷期/页码,标记卷年冲突、作者编造、页码偏差等 “verify refs”, “校验文献”, “check references”, “文献验证”, “ref check” 详情
nature-academic-search Beta 多源文献检索、引用核验、严格他引审计、文章引用指标表、高影响力引用者画像和参考文献管理 “search papers”, “find articles”, “literature search”, “查文献”, “verify DOI”, “严格他引”, “文章引用表”, “引用我的文章的人有没有大牛” 详情
nature-downloader Beta 通过图书馆资源入口、Chrome 登录态和开放获取路径合法获取学术全文/PDF “download papers”, “图书馆下载文献”, “CARSI”, “Web of Science”, “PDF 下载” 详情
nature-literature-pipeline Stable 自动化文献发现管线:多源检索、六维评分、精读推送和本地归档 “literature pipeline”, “每日文献”, “文献推送”, “daily literature push”, “cron” 详情
nature-experiment-log Draft 标准化记录实验图片、语音和文字材料,生成带 YAML frontmatter 的 Obsidian 实验日志并归档原始材料 “实验日志”, “记录实验”, “experiment log”, “Obsidian vault”, “飞书科研群” 详情
nature-proposal-writer Beta proposal-first 科研写作状态机,先建立证据、论证和章节契约,再起草或审查文本 “researchwrite”, “proposal”, “开题报告”, “研究方案”, “科研写作 QA” 详情

7. 贡献与开发

7.1 共享设计原则

所有技能都遵守以下原则:

  1. 优先使用一手来源:规则基于已发表 Nature 内容、官方期刊指南或明确的本地来源,而不是泛泛审美偏好。
  2. 显式胜过隐式:每条规则都应说明理由,而不是只给断言。
  3. 感知章节与任务上下文:学术写作、图件、引用和回复都依赖上下文;不同论文部分使用不同逻辑。
  4. 输出优先:每个技能都应返回能直接使用的产物,例如可粘贴文本、.svg.pptx.docx 或具体建议。
  5. 可扩展:每个技能自包含在自己的目录中,新增技能不应要求修改既有技能。

7.2 仓库目录结构

skills/
├── nature-shared/              # 当技能引用 ../nature-shared 时需要保留
├── nature-<topic>/
│   ├── README.md
│   ├── README_EN.md
│   ├── SKILL.md
│   ├── manifest.yaml     # router-style 技能会包含
│   ├── static/           # router-style 技能会包含
│   └── references/...
└── nature-proposal-writer/
    ├── README.md
    ├── README_EN.md
    ├── SKILL.md
    ├── scripts/...
    ├── templates/...
    └── references/...

7.3 新增技能流程

向本仓库添加技能时,请按以下流程:

1. 创建技能目录

skills/nature-<topic>/

2. 添加必需文件

文件 是否必需 用途
SKILL.md 必需 frontmatter(namedescription)+ 规则 + 工作流;触发后由 agent 加载
README.md 必需 面向人的中文说明文档
README_EN.md 必需 与中文详情页配套的英文说明文档
references/*.md 复杂技能推荐 模块化规则文件,例如 API、设计理论、教程、图表类型等

3. 编写中英文 README

每个新增技能都必须同时提供 README.mdREADME_EN.md。README 是面向人的技能入口页,不是 SKILL.md 的重复版,也不是安装手册。它的目标是让用户在 30 秒内判断:这个 skill 能不能解决我的问题、我要给它什么、它会产出什么、边界在哪里。

基本规则:

  • 中文 README 和英文 README 必须一一镜像:标题数量一致、顺序一致、信息点一致;英文页不要写成另一套独立模板。
  • 顶部固定为技能名、语言切换链接和一句定位说明。
  • 默认使用下面的基础结构;只有确实需要时才插入可选章节。
  • 不要在单个 skill README 中重复仓库安装教程、作者信息、变更日志、开发过程、完整文件树或大段内部实现细节。
  • 复杂规则、API 参数、长教程、脚本说明和模板索引应放进 references/static/scripts/SKILL.md,README 只保留路标。
  • 如果技能有视觉资产,可以放一个小型预览表;不要把 README 变成大型图库或长篇技术手册。

中文 README 基础结构:

# `nature-<topic>` 技能

[English](README_EN.md)

一句话说明这个技能的定位、主要任务和使用边界。

## 适合用它做什么
## 典型请求
## 你需要提供
## 产出
## 边界
## 相关技能

英文 README 必须对应为:

# `nature-<topic>` Skill

[中文说明](README.md)

One sentence describing the skill's role, main task, and usage boundary.

## What To Use It For
## Typical Requests
## What You Need To Provide
## Outputs
## Boundaries
## Related Skills

可选章节必须中英文同步插入,并保持相同顺序。常见可选章节包括:

中文章节 英文章节 使用场景
## 工作方式 ## Workflow 需要解释核心流程或路由方式
## 运行和依赖 ## Runtime and Dependencies 有脚本、MCP、API key、本地配置或外部依赖
## 示例预览 ## Example Preview 有少量图件、截图或可视化资产值得展示
## 内置参考 ## Built-In References 需要指向 references/assets/ 或 demo
## 方法来源 ## Method Sources 写作、审查或分析规则来自特定来源
## 三种模式 ## Three Modes 技能有清晰的 compose/revise/hybrid 等模式
## 与 ... 的关系 ## Relationship With ... 容易和另一个技能混淆,需要说明分工

提交前至少做这些 README 检查:

python scripts/validate-readmes.py
git diff --check
for d in skills/nature-*; do
  [ -f "$d/README.md" ] && [ -f "$d/README_EN.md" ] || continue
  rg -q '^\[English\]\(README_EN\.md\)$' "$d/README.md"
  rg -q '^\[中文说明\]\(README\.md\)$' "$d/README_EN.md"
  test "$(rg -c '^## ' "$d/README.md")" = "$(rg -c '^## ' "$d/README_EN.md")"
done

4. 录制使用教程

提交 PR 时,请同时录制一个简短的使用教程,说明这个 skill 解决什么问题、如何触发、需要什么输入,以及会产出什么结果。可以在 PR 描述中附上视频、录屏链接或可公开访问的教程地址。

5. 配置 SKILL.md frontmatter

---
name: nature-<topic>
description: >-
  用一句话说明这个技能做什么、什么时候触发、主要输出格式和核心使用场景。
---

6. 更新技能索引

在上方 技能索引 表格中添加一行:

| [`nature-<topic>`](skills/nature-<topic>/README.md) | Draft / Stable | 一句话用途 | 触发词 | [详情](skills/nature-<topic>/README.md) |

7. 设置状态标签

标签 含义
Draft 规则已定义,但尚未在真实案例上测试
Beta 已在示例上测试,仍可能存在边界问题
Stable 已在真实学术内容上验证,规则相对稳定

8. Star 历史

Star History Chart

View on GitHub

Recent activity

commits and pull requests

Code frequency

additions and deletions
+30.4K-30.4KWeek of 2026-04-19: +609 linesWeek of 2026-04-19: -91 linesWeek of 2026-04-26: +5,677 linesWeek of 2026-04-26: -1,765 linesWeek of 2026-05-03: +14,140 linesWeek of 2026-05-03: -8,102 linesWeek of 2026-05-10: +16,742 linesWeek of 2026-05-10: -397 linesWeek of 2026-05-17: +286 linesWeek of 2026-05-17: -88 linesWeek of 2026-05-24: +25,558 linesWeek of 2026-05-24: -2,316 linesWeek of 2026-05-31: +5,614 linesWeek of 2026-05-31: -2,594 linesWeek of 2026-06-07: +998 linesWeek of 2026-06-07: -34 linesWeek of 2026-06-14: +11,435 linesWeek of 2026-06-14: -30,441 linesWeek of 2026-06-21: +4,394 linesWeek of 2026-06-21: -2 linesWeek of 2026-06-28: +4,671 linesWeek of 2026-06-28: -5,337 linesWeek of 2026-07-05: +4,732 linesWeek of 2026-07-05: -329 linesWeek of 2026-07-12: +21,650 linesWeek of 2026-07-12: -4,025 linesWeek of 2026-07-19: +5,689 linesWeek of 2026-07-19: -1,641 linesWeek of 2026-07-26: +32 linesWeek of 2026-07-26: -33 linesApr 19, 2026Jul 26, 2026
+122.2K lines added, -57.2K removed over the last year.

Commits per week

last 52 weeks
1030Week of 2025-08-03: 0 commitsWeek of 2025-08-10: 0 commitsWeek of 2025-08-17: 0 commitsWeek of 2025-08-24: 0 commitsWeek of 2025-08-31: 0 commitsWeek of 2025-09-07: 0 commitsWeek of 2025-09-14: 0 commitsWeek of 2025-09-21: 0 commitsWeek of 2025-09-28: 0 commitsWeek of 2025-10-05: 0 commitsWeek of 2025-10-12: 0 commitsWeek of 2025-10-19: 0 commitsWeek of 2025-10-26: 0 commitsWeek of 2025-11-02: 0 commitsWeek of 2025-11-09: 0 commitsWeek of 2025-11-16: 0 commitsWeek of 2025-11-23: 0 commitsWeek of 2025-11-30: 0 commitsWeek of 2025-12-07: 0 commitsWeek of 2025-12-14: 0 commitsWeek of 2025-12-21: 0 commitsWeek of 2025-12-28: 0 commitsWeek of 2026-01-04: 0 commitsWeek of 2026-01-11: 0 commitsWeek of 2026-01-18: 0 commitsWeek of 2026-01-25: 0 commitsWeek of 2026-02-01: 0 commitsWeek of 2026-02-08: 0 commitsWeek of 2026-02-15: 0 commitsWeek of 2026-02-22: 0 commitsWeek of 2026-03-01: 0 commitsWeek of 2026-03-08: 0 commitsWeek of 2026-03-15: 0 commitsWeek of 2026-03-22: 0 commitsWeek of 2026-03-29: 0 commitsWeek of 2026-04-05: 0 commitsWeek of 2026-04-12: 0 commitsWeek of 2026-04-19: 9 commitsWeek of 2026-04-26: 23 commitsWeek of 2026-05-03: 66 commitsWeek of 2026-05-10: 103 commitsWeek of 2026-05-17: 71 commitsWeek of 2026-05-24: 100 commitsWeek of 2026-05-31: 32 commitsWeek of 2026-06-07: 11 commitsWeek of 2026-06-14: 46 commitsWeek of 2026-06-21: 2 commitsWeek of 2026-06-28: 21 commitsWeek of 2026-07-05: 18 commitsWeek of 2026-07-12: 76 commitsWeek of 2026-07-19: 18 commitsWeek of 2026-07-26: 4 commitsAug 3, 2025Jul 26, 2026
600 commits in the last 52 weeks.

When work happens

weekday and hour
SunMonTueWedThuFriSat036912151821Sun 0:00 — 5 commitsSun 1:00 — 5 commitsSun 2:00 — 2 commitsSun 3:00 — 1 commitsSun 4:00 — 3 commitsSun 5:00 — 1 commitsSun 6:00 — 3 commitsSun 7:00 — 2 commitsSun 8:00 — 2 commitsSun 9:00 — 4 commitsSun 10:00 — 4 commitsSun 11:00 — 6 commitsSun 12:00 — 3 commitsSun 13:00 — 4 commitsSun 14:00 — 1 commitsSun 15:00 — 6 commitsSun 16:00 — 8 commitsSun 17:00 — 5 commitsSun 18:00 — 4 commitsSun 19:00 — 13 commitsSun 20:00 — 2 commitsSun 21:00 — 12 commitsSun 22:00 — 5 commitsSun 23:00 — 5 commitsMon 0:00 — 4 commitsMon 1:00 — 1 commitsMon 2:00 — 0 commitsMon 3:00 — 2 commitsMon 4:00 — 3 commitsMon 5:00 — 1 commitsMon 6:00 — 0 commitsMon 7:00 — 0 commitsMon 8:00 — 3 commitsMon 9:00 — 1 commitsMon 10:00 — 2 commitsMon 11:00 — 1 commitsMon 12:00 — 7 commitsMon 13:00 — 1 commitsMon 14:00 — 3 commitsMon 15:00 — 1 commitsMon 16:00 — 4 commitsMon 17:00 — 2 commitsMon 18:00 — 3 commitsMon 19:00 — 3 commitsMon 20:00 — 3 commitsMon 21:00 — 4 commitsMon 22:00 — 4 commitsMon 23:00 — 6 commitsTue 0:00 — 5 commitsTue 1:00 — 2 commitsTue 2:00 — 5 commitsTue 3:00 — 1 commitsTue 4:00 — 3 commitsTue 5:00 — 0 commitsTue 6:00 — 1 commitsTue 7:00 — 1 commitsTue 8:00 — 2 commitsTue 9:00 — 5 commitsTue 10:00 — 1 commitsTue 11:00 — 1 commitsTue 12:00 — 2 commitsTue 13:00 — 1 commitsTue 14:00 — 1 commitsTue 15:00 — 2 commitsTue 16:00 — 2 commitsTue 17:00 — 7 commitsTue 18:00 — 1 commitsTue 19:00 — 4 commitsTue 20:00 — 4 commitsTue 21:00 — 5 commitsTue 22:00 — 6 commitsTue 23:00 — 2 commitsWed 0:00 — 3 commitsWed 1:00 — 0 commitsWed 2:00 — 0 commitsWed 3:00 — 0 commitsWed 4:00 — 3 commitsWed 5:00 — 0 commitsWed 6:00 — 0 commitsWed 7:00 — 1 commitsWed 8:00 — 3 commitsWed 9:00 — 0 commitsWed 10:00 — 3 commitsWed 11:00 — 2 commitsWed 12:00 — 4 commitsWed 13:00 — 4 commitsWed 14:00 — 6 commitsWed 15:00 — 7 commitsWed 16:00 — 3 commitsWed 17:00 — 5 commitsWed 18:00 — 4 commitsWed 19:00 — 6 commitsWed 20:00 — 6 commitsWed 21:00 — 4 commitsWed 22:00 — 10 commitsWed 23:00 — 5 commitsThu 0:00 — 5 commitsThu 1:00 — 0 commitsThu 2:00 — 5 commitsThu 3:00 — 1 commitsThu 4:00 — 3 commitsThu 5:00 — 1 commitsThu 6:00 — 1 commitsThu 7:00 — 2 commitsThu 8:00 — 1 commitsThu 9:00 — 0 commitsThu 10:00 — 6 commitsThu 11:00 — 1 commitsThu 12:00 — 3 commitsThu 13:00 — 7 commitsThu 14:00 — 3 commitsThu 15:00 — 7 commitsThu 16:00 — 3 commitsThu 17:00 — 7 commitsThu 18:00 — 4 commitsThu 19:00 — 5 commitsThu 20:00 — 4 commitsThu 21:00 — 10 commitsThu 22:00 — 5 commitsThu 23:00 — 4 commitsFri 0:00 — 3 commitsFri 1:00 — 9 commitsFri 2:00 — 0 commitsFri 3:00 — 2 commitsFri 4:00 — 2 commitsFri 5:00 — 1 commitsFri 6:00 — 1 commitsFri 7:00 — 2 commitsFri 8:00 — 2 commitsFri 9:00 — 2 commitsFri 10:00 — 2 commitsFri 11:00 — 15 commitsFri 12:00 — 8 commitsFri 13:00 — 3 commitsFri 14:00 — 2 commitsFri 15:00 — 11 commitsFri 16:00 — 9 commitsFri 17:00 — 8 commitsFri 18:00 — 2 commitsFri 19:00 — 2 commitsFri 20:00 — 7 commitsFri 21:00 — 6 commitsFri 22:00 — 3 commitsFri 23:00 — 5 commitsSat 0:00 — 4 commitsSat 1:00 — 1 commitsSat 2:00 — 1 commitsSat 3:00 — 3 commitsSat 4:00 — 2 commitsSat 5:00 — 2 commitsSat 6:00 — 5 commitsSat 7:00 — 1 commitsSat 8:00 — 1 commitsSat 9:00 — 3 commitsSat 10:00 — 4 commitsSat 11:00 — 3 commitsSat 12:00 — 5 commitsSat 13:00 — 3 commitsSat 14:00 — 2 commitsSat 15:00 — 5 commitsSat 16:00 — 16 commitsSat 17:00 — 6 commitsSat 18:00 — 3 commitsSat 19:00 — 6 commitsSat 20:00 — 5 commitsSat 21:00 — 6 commitsSat 22:00 — 6 commitsSat 23:00 — 5 commits
Commit volume by weekday and hour (UTC). Larger dots mean more commits.

Who is committing

last 52 weeks
Maintainer commits197 (27%)
Community commits524 (73%)

721 commits in total over the last year.

DateListRankStars gained
Jul 8, 2026daily#22+4
May 19, 2026daily#25+29
May 6, 2026daily#24+59