Skip to content
Back to search

researchclaw

上朝式科研:AI-powered research workflow showcase

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/alphalab-ustc--researchclaw/badge.svg)](/listings/alphalab-ustc--researchclaw)

HTML

<a href="/listings/alphalab-ustc--researchclaw"><img src="/api/listings/alphalab-ustc--researchclaw/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

📡 每日论文推荐

推荐今日论文

> 📡 今日论文推荐 | Daily Paper Scout > ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ > 🗓️ 2026-03-26 | 匹配兴趣: LLM · RL · Agentic AI > > 1️⃣ **RLVR: Reinforcement Learning with Verifiable Rewards** > 👤 Zhang et al. (2026) 🏷️ cs.LG · cs.AI > 💡 通过可验证奖励信号显著提升 LLM 数学推理能力 > 🎯 相关原因: 直接结合 RL + LLM reasoning,命中核心方向 > 🔗 https://arxiv.org/abs/2503.XXXXX > ...

---

📝 深度速读 + Deep Note 笔记

帮我读一下这篇:https://arxiv.org/abs/2503.19823

> 📝 Deep Note 完成 | AutoRefine > ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ > 📌 **AutoRefine: Search and Refine During Think** > 👤 Shi et al. | 📅 2025 | ⭐ 4/5 > 🔗 https://arxiv.org/abs/2503.19823 > > 💡 **核心发现:** 在推理链中插入知识精炼步骤比单纯增加检索更有效 > 📊 **关键数据:** 2WikiMHQA EM 67.7, HotpotQA EM 55.3 > ✨ **启发:** 精炼检索到的知识比检索本身更重要 > > 💾 笔记 → papers/2026-04-01_autorefine.md

想同时生成 HTML 页面?加 `--html`:

DNL https://arxiv.org/abs/2503.19823 --html

---

📋 管理论文列表

加入待读 https://arxiv.org/abs/2503.19823

> ✅ 已加入待读:**AutoRefine** | 共 7 篇待读

我的论文列表

> 📋 阅读列表已更新 → 打开 reading-list.html 查看看板 > To Read: 5 篇 · Reading: 2 篇 · Done: 12 篇

---

💡 灵感生成

给我一些研究灵感

> 💡 Research Idea 1: **Adaptive Memory Retrieval for RL Agents** > 📋 Problem: 当前 RL agent 的记忆检索策略是静态的,无法根据任务难度动态调整... > 🔧 Approach: 在 rollout 过程中,用 uncertainty estimation 决定何时检索记忆... > 📚 Related: AutoRefine (推理时检索) · MemOCR (记忆布局感知) > 🎯 Feasibility: 数据集现成,4×A100 × 3天,风险:检索延迟影响 rollout 速度

---

📄 论文写作

论文大纲:我想做一个 RL + Memory 的工作,针对 long-horizon agentic 任务

> 📄 论文大纲 | Paper Outline > 📌 Working Title: MemRL: Memory-Augmented Reinforcement Learning for Long-Horizon Agents > 🎯 Core Claim: 动态记忆检索机制显著提升 RL agent 在长程任务上的成功率 > 📍 Target Venue: NeurIPS 2026 > > ## 1. Introduction ... > ## 2. Related Work ... > (完整大纲展开)

rebuttal
[Reviewer 2: The comparison with baseline X is unfair because...]

> 📝 Rebuttal Draft > **Reviewer 2** > > R2.1: "The comparison with baseline X is unfair..." > **Response:** We thank the reviewer for this concern. The comparison is fair because...

---

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

SKILL.md · Manifest · 58 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. ---
  2. name: research-claw
  3. description: >
  4. Complete AI research assistant with 6 core capabilities: paper discovery, deep reading notes,
  5. reading list management, research taste learning, idea generation, and paper writing.
  6. Trigger words: 论文推荐, paper recommendation, arXiv, 读论文, paper notes, DNL,
  7. 推荐今日论文, 每日论文, 帮我读一下, paper scout, 论文速读, 帮我安装 ResearchClaw,
  8. research claws, researchclaw, 加入待读, 我的论文列表, reading list, 标记已读,
  9. 更新我的研究画像, 我的研究画像, research profile, 给我一些研究灵感, idea generator,
  10. 跨论文分析, 帮我写论文, paper writing, 论文大纲, 写引言, rebuttal.
  11. metadata:
  12. openclaw:
  13. homepage: https://github.com/AlphaLab-USTC/ResearchClaw
  14. version: "3.1.0"
  15. author: AlphaLab-USTC
  16. ---
  17. # ResearchClaw v3.1 — Modular Research Assistant
  18. ResearchClaw is a complete AI research companion with **6 atomic skills**.
  19. Each skill is self-contained and can be used independently.
  20. ## Skills
  21. | # | Skill | Directory | What it does |
  22. |---|---|---|---|
  23. | 1 | **Paper Scout** 📡 | `skills/paper-scout/` | Daily arXiv paper discovery with personalized ranking |
  24. | 2 | **Paper Reader** 📝 | `skills/paper-reader/` | Deep reading → markdown deep notes (default) or HTML page |
  25. | 3 | **Reading List** 📋 | `skills/reading-list/` | Kanban paper management + HTML dashboard |
  26. | 4 | **Research Profile** 🧠 | `skills/research-profile/` | Research taste profile + visualization |
  27. | 5 | **Idea Generator** 💡 | `skills/idea-generator/` | Cross-paper insights → research idea proposals |
  28. | 6 | **Paper Writer** 📄 | `skills/paper-writer/` | Outline → draft → auto-review → rebuttal |
  29. ## Quick Reference
  30. | Say this | Skill triggered |
  31. |---|---|
  32. | `推荐今日论文` | Paper Scout |
  33. | `帮我读一下 [arXiv link]` | Paper Reader |
  34. | `加入待读 [link]` | Reading List |
  35. | `我的论文列表` | Reading List |
  36. | `我的研究画像` | Research Profile |
  37. | `给我一些研究灵感` | Idea Generator |
  38. | `论文大纲 [idea]` | Paper Writer |
  39. | `rebuttal [comments]` | Paper Writer |
  40. ## Shared Config
  41. - **Research profile:** `~/.openclaw/workspace/research-claw-config.md`
  42. - **Reading list data:** `~/.openclaw/workspace/research-claw-reading-list.json`
  43. - **HTML output:** `~/.openclaw/workspace/research-claw-output/`
  44. - **Templates:** `templates/` (paper-note, reading-list, research-profile)
  45. ## Install
  46. ```
  47. 帮我安装 ResearchClaw:https://raw.githubusercontent.com/AlphaLab-USTC/ResearchClaw/main/docs/install.md
  48. ```
README.md · README · 388 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. <div align="center">
  2. # 🦞 ResearchClaw
  3. **你的 AI 全能论文助手 | Your Complete AI Research Companion**
  4. *由 [OpenClaw](https://openclaw.ai) 驱动 · Powered by OpenClaw*
  5. [![OpenClaw Skill](https://img.shields.io/badge/OpenClaw-Skill-blue?style=flat-square)](https://openclaw.ai)
  6. [![Version](https://img.shields.io/badge/version-3.1.0-brightgreen?style=flat-square)](https://github.com/AlphaLab-USTC/ResearchClaw)
  7. [![arXiv](https://img.shields.io/badge/arXiv-Free%20API-red?style=flat-square)](https://arxiv.org)
  8. [![License: MIT](https://img.shields.io/badge/License-MIT-yellow?style=flat-square)](LICENSE)
  9. [![Demo](https://img.shields.io/badge/Demo-GitHub%20Pages-green?style=flat-square)](https://alphalab-ustc.github.io/ResearchClaw/)
  10. <br>
  11. **6 大核心能力,零 API Key,开箱即用的 AI 科研全流程助手。**
  12. *6 core capabilities. No API keys. End-to-end AI research companion, out of the box.*
  13. <br>
  14. [🌐 在线演示](https://alphalab-ustc.github.io/ResearchClaw/) · [📖 安装](#一句话安装) · [⚙️ 配置](docs/config.md) · [🐛 Issues](https://github.com/AlphaLab-USTC/ResearchClaw/issues)
  15. </div>
  16. ---
  17. ## ✨ 6 大核心能力 | 6 Core Capabilities
  18. <table>
  19. <tr>
  20. <td width="33%" valign="top">
  21. ### 📡 论文雷达
  22. **Paper Scout**
  23. 每天从 arXiv 自动发现最新论文,按你的研究画像智能排序,推送 Top 5–10。
  24. - 🔍 覆盖 LLM/RL/多模态等方向
  25. - 🎯 个性化相关度评分
  26. - 📈 从你的反馈中自动学习
  27. - ⏰ 支持每日定时推送
  28. **触发:** `推荐今日论文`
  29. [▶ 查看演示](https://alphalab-ustc.github.io/ResearchClaw/showcase/case1-daily-signal-brief.html)
  30. </td>
  31. <td width="33%" valign="top">
  32. ### 📝 深度速读
  33. **Paper Reader**
  34. 发一个 arXiv 链接,AI 读完论文,基于 **Deep Note 7-section 框架** 生成结构化阅读笔记。
  35. - 📖 **Deep Note 框架**:Metadata → Why-read → CRGP → Figures → Experiments → Insights → Next steps
  36. - 📊 含图表 URL、具体实验数字、消融分析
  37. - 📝 **默认输出 Markdown**(git-friendly,低 token 消耗)
  38. - 🎨 可选生成 HTML 页面(`--html`)
  39. - 🇨🇳 中英双语混合
  40. **触发:** arXiv 链接 / `帮我读一下` / `DNL`
  41. [▶ 查看演示](https://alphalab-ustc.github.io/ResearchClaw/showcase/demo-paper-note.html)
  42. </td>
  43. <td width="33%" valign="top">
  44. ### 📋 论文列表
  45. **Reading List**
  46. 维护你的论文待读/正在读/已读列表,一键生成 HTML 看板。
  47. - 📊 三栏看板:To Read / Reading / Done
  48. - ➕ 一句话加入待读
  49. - ✅ 一句话标记已读
  50. - 🔗 每篇都链接到 HTML 笔记
  51. **触发:** `我的论文列表` / `加入待读`
  52. [▶ 查看演示](https://alphalab-ustc.github.io/ResearchClaw/showcase/demo-reading-list.html)
  53. </td>
  54. </tr>
  55. <tr>
  56. <td width="33%" valign="top">
  57. ### 🧠 研究画像
  58. **Research Taste Profile**
  59. 学习你的研究偏好,生成可视化研究画像页面,个性化一切推荐。
  60. - 📝 研究方向描述
  61. - 🌱 种子论文 & 关键词
  62. - 👤 关注作者白名单
  63. - 🔄 从对话反馈自动更新
  64. **触发:** `我的研究画像` / `更新我的研究画像`
  65. [▶ 查看演示](https://alphalab-ustc.github.io/ResearchClaw/showcase/demo-research-profile.html)
  66. </td>
  67. <td width="33%" valign="top">
  68. ### 💡 灵感生成器
  69. **Idea Generator**
  70. 分析你已读的论文,找出研究空白,生成 3–5 个有可行性的研究灵感。
  71. - 🔗 跨论文连接分析
  72. - 🕳️ 自动识别研究缺口
  73. - 📋 每个灵感含问题陈述 + 方法 + 相关论文 + 可行性评估
  74. - ⭐ 时机分析
  75. **触发:** `给我一些研究灵感`
  76. [▶ 查看演示](https://alphalab-ustc.github.io/ResearchClaw/showcase/case3-research-proposal.html)
  77. </td>
  78. <td width="33%" valign="top">
  79. ### 📄 论文写作
  80. **Paper Writer**
  81. 从灵感到草稿到回复审稿,覆盖论文写作全流程。
  82. - 🗂️ 结构化大纲生成
  83. - ✍️ 逐章节起草
  84. - 🔍 自我审稿 & 找弱点
  85. - 💬 Reviewer 回复起草
  86. **触发:** `论文大纲` / `写引言` / `rebuttal`
  87. [▶ 查看演示](https://alphalab-ustc.github.io/ResearchClaw/showcase/case4-multi-agent-codebase.html)
  88. </td>
  89. </tr>
  90. </table>
  91. ---
  92. ## 🚀 一句话安装
  93. 把这句话发给你的 OpenClaw 代理:
  94. ```
  95. 帮我安装 ResearchClaw:https://raw.githubusercontent.com/AlphaLab-USTC/ResearchClaw/main/docs/install.md
  96. ```
  97. > **就这一句话。** 代理会自动下载、注册技能,无需任何手动操作。
  98. **还没有 OpenClaw?** 先访问 [openclaw.ai](https://openclaw.ai) 安装,再运行上面这句话。
  99. ---
  100. ## 💬 快速上手示例 | Quick Start
  101. ### 📡 每日论文推荐
  102. ```
  103. 推荐今日论文
  104. ```
  105. > 📡 今日论文推荐 | Daily Paper Scout
  106. > ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  107. > 🗓️ 2026-03-26 | 匹配兴趣: LLM · RL · Agentic AI
  108. >
  109. > 1️⃣ **RLVR: Reinforcement Learning with Verifiable Rewards**
  110. > 👤 Zhang et al. (2026) 🏷️ cs.LG · cs.AI
  111. > 💡 通过可验证奖励信号显著提升 LLM 数学推理能力
  112. > 🎯 相关原因: 直接结合 RL + LLM reasoning,命中核心方向
  113. > 🔗 https://arxiv.org/abs/2503.XXXXX
  114. > ...
  115. ---
  116. ### 📝 深度速读 + Deep Note 笔记
  117. ```
  118. 帮我读一下这篇:https://arxiv.org/abs/2503.19823
  119. ```
  120. > 📝 Deep Note 完成 | AutoRefine
  121. > ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  122. > 📌 **AutoRefine: Search and Refine During Think**
  123. > 👤 Shi et al. | 📅 2025 | ⭐ 4/5
  124. > 🔗 https://arxiv.org/abs/2503.19823
  125. >
  126. > 💡 **核心发现:** 在推理链中插入知识精炼步骤比单纯增加检索更有效
  127. > 📊 **关键数据:** 2WikiMHQA EM 67.7, HotpotQA EM 55.3
  128. > ✨ **启发:** 精炼检索到的知识比检索本身更重要
  129. >
  130. > 💾 笔记 → papers/2026-04-01_autorefine.md
  131. 想同时生成 HTML 页面?加 `--html`:
  132. ```
  133. DNL https://arxiv.org/abs/2503.19823 --html
  134. ```
  135. ---
  136. ### 📋 管理论文列表
  137. ```
  138. 加入待读 https://arxiv.org/abs/2503.19823
  139. ```
  140. > ✅ 已加入待读:**AutoRefine** | 共 7 篇待读
  141. ```
  142. 我的论文列表
  143. ```
  144. > 📋 阅读列表已更新 → 打开 reading-list.html 查看看板
  145. > To Read: 5 篇 · Reading: 2 篇 · Done: 12 篇
  146. ---
  147. ### 💡 灵感生成
  148. ```
  149. 给我一些研究灵感
  150. ```
  151. > 💡 Research Idea 1: **Adaptive Memory Retrieval for RL Agents**
  152. > 📋 Problem: 当前 RL agent 的记忆检索策略是静态的,无法根据任务难度动态调整...
  153. > 🔧 Approach: 在 rollout 过程中,用 uncertainty estimation 决定何时检索记忆...
  154. > 📚 Related: AutoRefine (推理时检索) · MemOCR (记忆布局感知)
  155. > 🎯 Feasibility: 数据集现成,4×A100 × 3天,风险:检索延迟影响 rollout 速度
  156. ---
  157. ### 📄 论文写作
  158. ```
  159. 论文大纲:我想做一个 RL + Memory 的工作,针对 long-horizon agentic 任务
  160. ```
  161. > 📄 论文大纲 | Paper Outline
  162. > 📌 Working Title: MemRL: Memory-Augmented Reinforcement Learning for Long-Horizon Agents
  163. > 🎯 Core Claim: 动态记忆检索机制显著提升 RL agent 在长程任务上的成功率
  164. > 📍 Target Venue: NeurIPS 2026
  165. >
  166. > ## 1. Introduction ...
  167. > ## 2. Related Work ...
  168. > (完整大纲展开)
  169. ```
  170. rebuttal
  171. [Reviewer 2: The comparison with baseline X is unfair because...]
  172. ```
  173. > 📝 Rebuttal Draft
  174. > **Reviewer 2**
  175. > > R2.1: "The comparison with baseline X is unfair..."
  176. > **Response:** We thank the reviewer for this concern. The comparison is fair because...
  177. ---
  178. ## 🎬 在线演示 | Live Showcase
  179. 访问 [alphalab-ustc.github.io/ResearchClaw](https://alphalab-ustc.github.io/ResearchClaw/) 查看完整演示:
  180. | 案例 | 描述 |
  181. |------|------|
  182. | [每日信号简报](https://alphalab-ustc.github.io/ResearchClaw/showcase/case1-daily-signal-brief.html) | Paper Scout 多源聚合推送示例 |
  183. | [论文阅读笔记](https://alphalab-ustc.github.io/ResearchClaw/showcase/case2-paper-reading-notes.html) | Paper Reader HTML 笔记示例 |
  184. | [研究方向分析](https://alphalab-ustc.github.io/ResearchClaw/showcase/case3-research-proposal.html) | Idea Generator 灵感生成示例 |
  185. | [多智能体协作](https://alphalab-ustc.github.io/ResearchClaw/showcase/case4-multi-agent-codebase.html) | Paper Writer 写作辅助示例 |
  186. ---
  187. ## ⚙️ 配置研究画像
  188. 安装后,直接告诉代理:
  189. ```
  190. 更新我的研究画像:我主要研究强化学习和多模态 agent,关注 RLVR 方向
  191. ```
  192. 或手动编辑 `~/.openclaw/workspace/research-claw-config.md`:
  193. ```yaml
  194. research_direction: "PhD in LLM reasoning and agentic memory systems"
  195. seed_papers:
  196. - 2503.19823 # AutoRefine
  197. keywords:
  198. - reinforcement learning
  199. - agentic memory
  200. - long-context reasoning
  201. whitelist_authors:
  202. - Yaorui Shi
  203. - An Zhang
  204. ```
  205. 详见 [配置指南](docs/config.md)
  206. ---
  207. ## 📦 每日自动推送
  208. ```
  209. 帮我设置每天早上 9 点自动推荐论文
  210. ```
  211. 代理会自动创建 OpenClaw cron 任务。✅
  212. ---
  213. ## 🔧 工作原理 | How It Works
  214. ```
  215. 用户: 推荐今日论文
  216. ↓
  217. 读取 research-claw-config.md(研究画像)
  218. ↓
  219. 调用 arXiv 公开 API(免费,无需 Key)
  220. ↓
  221. 按个性化偏好评分排序
  222. ↓
  223. 返回 Top 5–10 论文速览
  224. ↓
  225. 用户反馈 → 自动更新研究画像
  226. ```
  227. ```
  228. 用户: 帮我读一下 arxiv.org/abs/XXXXX
  229. ↓
  230. 提取 arXiv ID → 获取元数据 + HTML 版本
  231. ↓
  232. 分析论文内容(Introduction, Method, Experiments)
  233. ↓
  234. 按 Deep Note 7-section 框架提取结构化信息
  235. ↓
  236. 输出 Markdown 笔记文件(默认)+ 聊天摘要
  237. ↓
  238. (可选)生成 HTML 页面
  239. ```
  240. **零第三方依赖**:只需 OpenClaw + 网络,不需要任何额外 API Key。
  241. ---
  242. ## 🤝 Credits
  243. - 🦞 Built with [OpenClaw](https://openclaw.ai) — AI assistant platform
  244. - 📄 Paper data from [arXiv](https://arxiv.org) (free open-access API)
  245. - 🧠 Created by [AlphaLab@USTC](https://github.com/AlphaLab-USTC) (Yaorui Shi et al.)
  246. ---
  247. ## 🌍 English Quick Reference
  248. **Install** (one line — send to your OpenClaw agent):
  249. ```
  250. 帮我安装 ResearchClaw:https://raw.githubusercontent.com/AlphaLab-USTC/ResearchClaw/main/docs/install.md
  251. ```
  252. **6 capabilities at a glance:**
  253. | Say this | What happens |
  254. |---|---|
  255. | `推荐今日论文` | Daily arXiv paper recommendations, ranked by your profile |
  256. | `帮我读一下 [arXiv link]` | Deep read → markdown deep note (+ optional HTML with `--html`) |
  257. | `加入待读 [link]` | Add paper to your reading list |
  258. | `我的论文列表` | View & regenerate your HTML reading dashboard |
  259. | `更新我的研究画像` | Update research taste profile + visual HTML page |
  260. | `给我一些研究灵感` | Cross-paper analysis → 3–5 research idea proposals |
  261. | `论文大纲 [idea]` | Generate structured paper outline |
  262. | `写引言` / `写方法` | Draft specific paper sections |
  263. | `rebuttal [comments]` | Draft point-by-point reviewer responses |
  264. **No API key needed.** Uses arXiv's free public API + OpenClaw's built-in tools.
  265. Zero third-party dependencies.
  266. ---
  267. ---
  268. ## 🔗 More from AlphaLab
  269. Check out our other projects:
  270. - 🦀 [**PaperClaw**](https://github.com/AkaliKong/PaperClaw) — Automated paper collection and analysis toolkit
  271. - 🛠️ [**OhMyCode**](https://github.com/AlphaLab-USTC/OhMyCode) — Code generation and repair benchmark
  272. - 📚 [**Must-Read-LLM-Papers**](https://github.com/AlphaLab-USTC/Must-Read-LLM-Papers) — Curated reading list for LLM research
  273. ---
  274. <div align="center">
  275. Made with ❤️ for researchers · MIT License · [alphalab-ustc.github.io/ResearchClaw](https://alphalab-ustc.github.io/ResearchClaw/)
  276. </div>

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.