⏳ This skill is pending AI review.

Scores will appear once the review pipeline completes.

version unknown

ai-premium-rs-papers

yuanshuai0914@yuanshuai0914⭐ 0 stars· last commit 7mo ago· 0 open issues

Fetch remote sensing papers from premium publishers: Nature, Science, and PNAS.

Choose how to use this skill

You do not need every option. Choose the path your AI client supports. The stable page stays the same; versioned files are immutable.

1. Native installer

This listing has no registered native installer command. Use the complete package or source fallback below, depending on what your client supports.

Do not guess an installer command or replace an existing version without reviewing the diff.

2. Complete package recommended

Download the ZIP when available. It includes SKILL.md plus the references, security notes and version metadata.

No complete ProSkills package is published for this listing yet.

3. Prompt-only

Copy the prompt above when the agent can read the stable page or when you want to adopt the workflow without installing a skill.

Need only the instruction file?

Download SKILL.md only if your client requires a single file. The complete ZIP is safer for a full installation because it preserves the references and release context.

No path installs or executes anything by itself. Your agent still needs access to the project files. Before updating, compare the installed version and review the diff.

—/10

// RATINGS

⭐GitHub Stars
⭐ 0 on GitHubGitHub ↗

New / niche

🟢ProSkills Score
—
📍

Not yet listed on ClawHub or SkillsMP

// README

airs-daily-paper

论文推荐、论文筛选、论文精读,最后自动存进本地 Markdown 知识库;同步到 Feishu。

🦴 这套东西会帮我做什么

  • 抓取每日新论文,初筛后生成推荐列表。
  • 支持完整解析、快速摘要和批判性分析。
  • 术语可沉淀为概念笔记,方便后续串联。
  • 自动写入本地 Markdown 知识库,并维护目录页和导航页。
  • 同步到 Feishu 文档。
  • 可接入 Zotero,省去手动复制链接。

最终生成结果在本地知识库里大概会长这样:

PaperKnowledgeBase/
├── DailyPapers/YYYY-MM-DD-论文推荐.md
├── 论文笔记/.../*.md
└── 论文笔记/_概念/.../*.md

🐕 怎么用

基本就 2 句:

今日论文推荐
读一下这篇论文 https://arxiv.org/abs/2509.24527

其他常见说法:

过去3天论文推荐
过去一周论文推荐

读一下这篇论文 ~/Downloads/paper.pdf
快速看一下这篇论文 https://arxiv.org/abs/2509.24527
批判性分析这篇论文 ~/Downloads/paper.pdf

读一下 Zotero 里的 Diffusion Policy
批量读一下 Zotero 里 VLA 分类下的论文

今日论文推荐 会跑完整流程,读一下这篇论文 ... 用来读单篇。

目录页一般会自动更新;如果你手动改过结构,或者怀疑没同步,再补一句:

更新索引

🏡 安装

前置环境:

建议给本地知识库目录加上 git 版本管理。笔记多了以后有个版本历史会安心很多,也方便多设备同步。

如果你是在自己的本地机器上日常使用,通常直接用 claude --dangerously-skip-permissions 会省很多权限确认;前提是你清楚这会跳过权限检查,所以更适合个人环境,不建议在不熟悉的机器上直接这么跑。

在仓库根目录运行:

mkdir -p ~/.claude/skills
cp -r ./skills/* ~/.claude/skills/

# 改成你自己的本地知识库路径,要跟配置文件里的 paths.knowledge_base_root 一致
VAULT=~/PaperKnowledgeBase
mkdir -p "$VAULT/DailyPapers" \
  "$VAULT/论文笔记/_概念/0-待分类" \
  "$VAULT/论文笔记/_待整理"

⚙️ 配置

安装完之后需要改一下配置。配置文件是 ~/.claude/skills/_shared/user-config.json,可以自己改,也可以直接告诉 Claude 你的需求让它帮你改。

里面主要改这几项:

配置项说明
paths.knowledge_base_root本地 Markdown 知识库路径
paths.obsidian_vault旧字段,兼容保留;没配 knowledge_base_root 时会回退到这里
paths.zotero_dbZotero 数据库路径(不用 Zotero 可以不填)
paths.zotero_storageZotero 附件存储路径
daily_papers.keywords你关心的研究方向,用来给论文打分
daily_papers.negative_keywords你不想看的方向,命中直接排除
daily_papers.domain_boost_keywords额外加分的领域词
publishing.backend发布后端:obsidian 或 feishu
publishing.auto_sync生成/更新 Markdown 后是否自动同步到 Feishu
publishing.feishu.cli_pathlark-cli 可执行文件路径
publishing.feishu.wiki_space / folder_tokenFeishu 文档的目标空间或目录

批量读一下 Zotero 里 XXX 分类下的论文 不需要额外的映射文件;只要 paths.zotero_db 和 paths.zotero_storage 配对,脚本会直接从你的 Zotero 分类树里查。

🦮 默认行为

默认本地知识库管理不会自动 commit、push:

  • auto_refresh_indexes = true
  • git_commit = false
  • git_push = false

也就是默认会自动刷新目录页,但不会动你的 git。如果你的本地知识库已经用 git 管理,希望跑完流程后自动提交,把 git_commit 打开就行。

Feishu 同步

{
  "paths": {
    "knowledge_base_root": "~/PaperKnowledgeBase"
  },
  "publishing": {
    "backend": "feishu",
    "auto_sync": true,
    "feishu": {
      "cli_path": "~/.npm-global/bin/lark-cli",
      "wiki_space": "my_library"
    }
  }
}

第一次使用前先装并登录 lark-cli:

npm install -g @larksuite/cli
lark-cli config init
lark-cli auth login --recommend

仓库里新增了 skills/_shared/feishu_sync.py。在 backend=feishu 且 auto_sync=true 时,相关 skill 会在本地 Markdown 落盘后调用这个脚本,把每日推荐、论文笔记和概念笔记同步成 Feishu 文档。

如果 lark-cli 不在你的 PATH 里,把 publishing.feishu.cli_path 改成它的绝对路径就行。

🐾 大概怎么跑的

每日推荐拆成三步流水线,避免单次上下文太长:

  1. 抓取:Python 脚本并发请求 HuggingFace Daily / Trending + arXiv API,按你配的关键词打分、去重,输出 top 30 候选到 /tmp。然后异步抓 arXiv 页面补全作者、机构、图片等元数据。
  2. 点评:Claude 读候选列表,按 必读 / 值得看 / 可跳过 分流,写锐评,保存到本地知识库的 DailyPapers/ 目录,同时更新 .history.json 做跨天去重。
  3. 笔记:对"必读"论文逐篇调 paper-reader 生成完整笔记(公式、图表、关键方法),顺便补概念库,最后回填链接、刷新目录页。

读单篇走 paper-reader:支持 arXiv 链接、本地 PDF、Zotero 搜索。会从 arXiv HTML / 项目主页 / PDF 多路取图,按模板生成结构化笔记,自动归类到本地知识库对应目录;如果启用了 Feishu,同步后会在飞书里生成对应文档。

目录页由 generate-mocs 维护:递归扫描论文笔记和概念库目录,自动生成带 wikilink 的索引页。

更多实现细节见 ARCHITECTURE.md。

🏠 仓库里有什么

平时真正常用的是前 2 个,后 1 个偏维护:

  • daily-papers:每日推荐全流程
  • paper-reader:读单篇论文
  • generate-mocs:手动补刷目录页

另外还有 3 个内部 skill,主要给调试和重跑单步用:

  • daily-papers-fetch
  • daily-papers-review
  • daily-papers-notes

🎾 进阶用法

如果你只想单独跑流水线某一步,也可以分别说:

跑一下论文抓取
跑一下论文点评
跑一下论文笔记

如果你想做本地定时任务(比如每天早上6点自动运行),可以直接让 Claude 按你的系统环境帮你配置。

🐶 FAQ

可以一步跑完整流程吗?

可以。直接说 今日论文推荐 就行。内部拆成三步主要是为了避免单次上下文过长,同时方便单步调试和重跑。

目录页会自动刷新吗?

默认会。读单篇论文和跑完整的每日推荐流程时,结束后通常都会自动刷新一次。更新索引 更像是手动补刷入口。

不用 Zotero 可以吗?

可以。每日推荐不依赖 Zotero,单篇阅读也支持直接输入 arXiv 链接或本地 PDF。Zotero 主要用于已有文献库的搜索、归类和批量处理。

不用 Obsidian 可以吗?

可以。输出本质上是 Markdown 文件,不强绑 Obsidian。现在也支持通过 lark-cli 同步到 Feishu;本地目录仍然会保留一份 Markdown 作为缓存和索引来源。

可以用来辅助论文写作吗?

可以,比较适合用来整理 related work、维护笔记库和生成阅读提纲。AI 生成的内容建议自己核验后再使用。

感谢

感谢 https://github.com/huangkiki/dailypaper-skills 提供的思路和模版。

// HOW IT'S BUILT

// FILES

2 files shipped with this skill. Inline text is the catalog copy. The upstream link is the source repository.

  • skills/ai-premium-rs-papers/SKILL.md4.6 KBupstream ↗
    # AI Premium Remote Sensing Papers
    
    Fetch remote sensing papers from premium publishers: Nature, Science, and PNAS.
    
    ## Overview
    
    This skill automatically fetches academic papers related to remote sensing, satellite monitoring, and earth observation from three prestigious publishers:
    
    - **Nature** - Leading scientific journal
    - **Science** - Premier AAAS journal  
    - **PNAS** - Proceedings of the National Academy of Sciences
    
    ## API Key Requirements
    
    | Source | API Key Required | Notes |
    |--------|------------------|-------|
    | Nature | Optional | RSS feed works without API key; API provides more results |
    | Science | Optional | RSS feed available, may have rate limits |
    | PNAS | No | Public RSS feeds available |
    
    ### API keys are optional - will use available sources only.
    
    ## Setup Instructions
    
    ### 1. Basic Setup (No API Keys)
    
    The skill works out-of-the-box using RSS feeds:
    
    ```bash
    cd ~/.openclaw/workspace/skills/ai-premium-rs-papers
    ./script.sh
    ```
    
    ### 2. With API Keys (Enhanced Results)
    
    #### Obtaining Nature API Key
    
    1. Visit [SpringerNature API Portal](https://dev.springernature.com/)
    2. Sign up for a free API key
    3. Copy your API key
    
    #### Obtaining Science API Key
    
    1. Visit [AAAS Developer Portal](https://www.science.org/about/
    2. Contact AAAS for API access if available
    3. Note: RSS feed is often sufficient
    
    #### Set Environment Variables
    
    ```bash
    export NATURE_API_KEY="your_nature_api_key"
    export SCIENCE_API_KEY="your_science_api_key"
    export FEISHU_TARGET="your_feishu_user_id"
    ```
    
    ### 3. Automated Daily Delivery
    
    #### Install LaunchAgent (macOS)
    
    1. Create logs directory:
    ```bash
    mkdir -p ~/.openclaw/workspace/skills/ai-premium-rs-papers/logs
    ```
    
    2. Copy and configure the plist:
    ```bash
    # Edit the plist file first to set your API keys and FEISHU_TARGET
    vim ~/.openclaw/workspace/skills/ai-premium-rs-papers/com.ai-premium-rs-papers.plist
    
    # Copy to LaunchAgents
    cp ~/.openclaw/workspace/skills/ai-premium-rs-papers/com.ai-premium-rs-papers.plist \
       ~/Library/LaunchAgents/
    
    # Load the agent
    launchctl load ~/Library/LaunchAgents/com.ai-premium-rs-papers.plist
    
    # Start immediately (optional)
    launchctl start com.ai-premium-rs-papers
    ```
    
    #### Schedule
    
    - **Runs daily at 08:30** (30 minutes after arXiv skill)
    - Generates report + sends top 5 papers to Feishu
    
    ### 4. Manual Run with Feishu Send
    
    ```bash
    export FEISHU_TARGET="your_feishu_user_id"
    ./run_and_send.sh
    ```
    
    ## File Structure
    
    ```
    skills/ai-premium-rs-papers/
    ├── generator.py              # Core fetching logic
    ├── script.sh                 # Manual execution
    ├── run_and_send.sh           # Generate + send to Feishu
    ├── com.ai-premium-rs-papers.plist  # LaunchAgent config
    ├── SKILL.md                  # This file
    └── recommendations/          # Generated reports
        └── YYYY-MM-DD.md
    ```
    
    ## Output Format
    
    ### Full Report (Markdown)
    
    Generated at `recommendations/YYYY-MM-DD.md`:
    
    ```markdown
    # Premium Remote Sensing Papers - 2024-01-15
    
    ## Summary
    Total papers found: 12
    
    | Source | Count |
    |--------|-------|
    | Nature | 5 |
    | PNAS | 7 |
    
    ### [Paper Title](url)
    - **Source:** Nature
    - **Authors:** Smith et al.
    - **Published:** 2024-01-15
    - **DOI:** 10.1038/...
    - **Abstract:** ...
    ```
    
    ### Feishu Summary (Compact)
    
    ```
    📡 每日遥感顶刊速递 (2024-01-15)
    
    1. **Paper Title** (Nature)
       👤 Smith, Jones et al. | 📅 2024-01-15
    
    2. **Another Title** (PNAS)
       👤 Wang et al. | 📅 2024-01-14
    
    ... 还有 7 篇论文 (详见完整报告)
    ```
    
    ## Configuration
    
    ### Environment Variables
    
    | Variable | Required | Description |
    |----------|----------|-------------|
    | `NATURE_API_KEY` | No | SpringerNature API key for enhanced results |
    | `SCIENCE_API_KEY` | No | AAAS API key (if available) |
    | `PNAS_API_KEY` | No | Not typically needed |
    | `FEISHU_TARGET` | For sending | Feishu user/chat ID |
    | `AI_PREMIUM_RS_PAPERS_PROXY` | No | HTTP proxy URL |
    
    ### Filter Keywords
    
    Papers are filtered for relevance to:
    - `remote sensing`
    - `satellite monitoring`
    - `earth observation`
    
    ## Troubleshooting
    
    ### No papers found
    
    - Check network connectivity
    - Verify RSS feeds are accessible
    - Some days may have no matching papers
    
    ### API errors
    
    - RSS fallback is used automatically
    - API keys may need renewal
    - Check rate limits
    
    ### Feishu send fails
    
    - Verify `FEISHU_TARGET` is set correctly
    - Ensure openclaw CLI is configured
    - Check Feishu permissions
    
    ## Dependencies
    
    - Python 3.6+
    - requests library (auto-installed)
    - openclaw CLI (for Feishu sending)
    
    ## License
    
    MIT - Open source, use freely.
    
    ## Updates
    
    To update the skill, modify the files in place. The LaunchAgent will use the updated version on next run.
    
  • README.mdupstream ↗

    Contents are not inlined in the catalog. Open the upstream file.

Plain-text copy, no login: /skill-bundles/yuanshuai0914-airs-daily-paper-skills-ai-premium-rs-papers.md

// REPO STATS

0 stars
0 open issues
Last commit: 7mo ago