Nature Reader
jing1312/nature-figure-skill
Build full-paper Chinese-English side-by-side, figure/table-aware, source-grounded Markdown readers for journal or conference papers from PDF, DOI, arXiv, publisher HTML, or pasted text.
Deep-read an arXiv paper and generate structured Deep Note reading notes.
Install Paper Reader
$ npx skills add AlphaLab-USTC/ResearchClaw --skill paper-reader -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install AlphaLab-USTC/ResearchClaw paper-reader --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/AlphaLab-USTC/ResearchClaw.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/paper-reader .claude/skills/paper-reader && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "paper-reader" agent skill from https://github.com/AlphaLab-USTC/ResearchClaw/tree/main/skills/paper-reader into .claude/skills/paper-reader/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "paper-reader", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/AlphaLab-USTC/ResearchClaw/tree/main/skills/paper-readerType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add AlphaLab-USTC/ResearchClaw --skill paper-reader -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install AlphaLab-USTC/ResearchClaw paper-reader --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/AlphaLab-USTC/ResearchClaw.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/paper-reader .agents/skills/paper-reader && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "paper-reader" agent skill from https://github.com/AlphaLab-USTC/ResearchClaw/tree/main/skills/paper-reader into .agents/skills/paper-reader/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "paper-reader", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add AlphaLab-USTC/ResearchClaw --skill paper-reader -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install AlphaLab-USTC/ResearchClaw paper-reader --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/AlphaLab-USTC/ResearchClaw.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/paper-reader .cursor/skills/paper-reader && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "paper-reader" agent skill from https://github.com/AlphaLab-USTC/ResearchClaw/tree/main/skills/paper-reader into .cursor/skills/paper-reader/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "paper-reader", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/AlphaLab-USTC/ResearchClaw.git --path skills/paper-reader--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add AlphaLab-USTC/ResearchClaw --skill paper-reader -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install AlphaLab-USTC/ResearchClaw paper-reader --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/AlphaLab-USTC/ResearchClaw.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/paper-reader .gemini/skills/paper-reader && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "paper-reader" agent skill from https://github.com/AlphaLab-USTC/ResearchClaw/tree/main/skills/paper-reader into .gemini/skills/paper-reader/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "paper-reader", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install AlphaLab-USTC/ResearchClaw paper-readerInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add AlphaLab-USTC/ResearchClaw --skill paper-reader -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/AlphaLab-USTC/ResearchClaw.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/paper-reader .github/skills/paper-reader && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "paper-reader" agent skill from https://github.com/AlphaLab-USTC/ResearchClaw/tree/main/skills/paper-reader into .github/skills/paper-reader/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "paper-reader", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add AlphaLab-USTC/ResearchClaw --skill paper-reader -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install AlphaLab-USTC/ResearchClaw paper-reader --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/AlphaLab-USTC/ResearchClaw.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/paper-reader .opencode/skills/paper-reader && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "paper-reader" agent skill from https://github.com/AlphaLab-USTC/ResearchClaw/tree/main/skills/paper-reader into .opencode/skills/paper-reader/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "paper-reader", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
paper-reader- GitHub stars
- 134
- Token cost
- ~1.6k tokens
- SKILL.md length
- 521 words
- Files
- 1
- Skills in repo
- 7
- Repo updated
- First seen
- Licence
- MIT
At a glance
Deep-read an arXiv paper and generate structured Deep Note reading notes.
- Works in 4 steps: Load Research Profile (Always First) → Extract arXiv ID → Fetch Metadata → …
- Tasks that involve Academic paper search
- SKILL.md covers Overview, Step 0 — Load Research Profile…, Step 1 — Extract arXiv ID and Step 2 — Fetch Metadata, plus 3 more sections
- Reaches arxiv.org
What it does
Paper Reader is an agent skill from AlphaLab-USTC/ResearchClaw. Deep-read an arXiv paper and generate structured Deep Note reading notes. Default output: markdown deep note file (low token cost, git-friendly). Optional: HTML page from template. Trigger: 帮我读一下, DNL, 论文速读, paper notes, or an arXiv link.
Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Research & Science, covering Academic paper search, HTML artifacts and LLM cost and token optimization. It works with arXiv and Git. The repository describes itself as: 上朝式科研：AI-powered research workflow showcase. The licence is MIT.
When your agent uses it
- Tasks that involve Academic paper search
- Tasks that involve HTML artifacts
- Tasks that involve LLM cost and token optimization
Example prompts
- “/paper-reader”
Workflow steps
4 steps, taken from the step headings in SKILL.md.
What it can do on your machine
Read from SKILL.md and the folder at commit 9d64c4b. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).
From the folder's file list and the shell code blocks in SKILL.md.
Network
Hosts in commands or code, which the agent is likely to contact:
arxiv.org
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Paper Reader loads about 1.6k tokens when it runs. Until then it costs about 63 tokens; SKILL.md has 521 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from AlphaLab-USTC/ResearchClaw at commit 9d64c4b, republished under its MIT licence (© AlphaLab-USTC). 521 words, ~1,599 tokens.
Download SKILL.mdSave it as
.claude/skills/paper-reader/SKILL.md (or your agent's skills folder).- name
- paper-reader
- description
- Deep-read an arXiv paper and generate structured Deep Note reading notes. Default output: markdown deep note file (low token cost, git-friendly). Optional: HTML page from template. Trigger: 帮我读一下, DNL, 论文速读, paper notes, or an arXiv link.
Paper Reader — Deep Reading Notes (v3.1)
Overview
Given an arXiv paper, produce a structured deep reading note as a markdown file. Optionally generate an HTML page from the paper-note template.
Default output format: Markdown Deep Note (saves tokens, git-friendly, composable).
HTML output: Add --html or say 生成 HTML to also produce an HTML page.
Triggers: arXiv link · 帮我读一下 · DNL · 论文速读 · paper notes
Step 0 — Load Research Profile (Always First)
Before running, load: ~/.openclaw/workspace/research-claw-config.md
If missing, use defaults silently and mention at the end:
Want to customize? Say "更新我的研究画像"
Step 1 — Extract arXiv ID
Recognize patterns:
https://arxiv.org/abs/2503.XXXXX→2503.XXXXXhttps://arxiv.org/pdf/2503.XXXXX→2503.XXXXXhttps://arxiv.org/pdf/2503.XXXXX.pdf→2503.XXXXX- Bare ID:
2503.XXXXX
Step 2 — Fetch Metadata
web_fetch: https://arxiv.org/abs/{ARXIV_ID}Extract: title, authors, year, abstract, subject categories, venue if mentioned.
Also fetch HTML version for figures: https://arxiv.org/html/{ARXIV_ID}v1
---## Step 3 — Analyze Paper Content
Use web_fetch on the HTML version for structured content extraction.
Focus on: Abstract, Introduction, Method, Experiments (tables/numbers), Conclusion.
Extract using the Deep Note 7-section framework:
| Section | What to extract |
|---|---|
| 0) Metadata | Title, alias, authors, venue, date, links, tags, rating, scoring breakdown |
| 1) Why-read | One-sentence: key claim + key observation |
| 2) CRGP | Context, Related work, Gap, Proposal — from Introduction |
| 3) Figures | Key figures with URLs from arxiv HTML + one-line descriptions |
| 4) Experiments | Main results table, ablation highlights, limitations |
| 5) Why it matters | Research insights for the reader's own work (2-4 bullets) |
| 6) Next steps | Actionable follow-up items as checkboxes |
| 7) Scoring | Rating breakdown explanation |
Scoring System
Base score: 1 (any complete paper with benchmarks)
Quality bonus (0-2):
- +1: Solid experiments with proper ablation
- +2: Strong ablation + SOTA results + novel methodology
Observation bonus (0-2):
- +1: Finding directly relevant to reader's research
- +2: Paradigm-shifting insight for the field
Final = Base + Quality + Observation (max 5/5)
---## Step 4 — Write Markdown Deep Note File
Output directory: Same repo as reading notes (e.g., papers/ directory).
Filename: YYYY-MM-DD_{alias}.md (e.g., 2026-04-01_gems.md)
Markdown Deep Note Template
# Deep Note — {ALIAS}
## 0) Metadata
- **Title:** {FULL_TITLE}
- **Alias:** {ALIAS}
- **Authors / Org:** {AUTHORS} ({INSTITUTIONS})
- **Venue / Status:** {VENUE_OR_ARXIV_ID} ({STATUS})
- **Date:** {PAPER_DATE}
- **Links:**
- Abs: https://arxiv.org/abs/{ARXIV_ID}
- HTML: https://arxiv.org/html/{ARXIV_ID}v1
- PDF: https://arxiv.org/pdf/{ARXIV_ID}
- Code: {CODE_URL_OR_PROJECT_PAGE}
- **Tags:** {comma-separated tags}
- **My rating:** {STARS} ({N}/5)
- **Read depth:** deep
- **Scoring ({BREAKDOWN}):** {EXPLANATION} = **{N}/5**
---
## 1) 一句话 Why-read
- **Key claim/contribution + key observation：** {ONE_PARAGRAPH}
---
## 2) CRGP 拆解 Introduction
### C — Context
{2-3 sentences on research background}
### R — Related work
{Bullet list grouped by methodology line}
### G — Gap
{2-3 sentences on specific limitations}
### P — Proposal
{2-3 sentences on proposed solution + key insight}
---
## 3) Figure 区
{For each key figure:}
- 图N（{description}）：
{One-line interpretation}
---
## 4) Experiments — Key Numbers
### Main Results
| Benchmark | Metric | This Work | Best Baseline | Delta |
|-----------|--------|-----------|---------------|-------|
| ... | ... | ... | ... | ... |
### Ablation
{Key ablation findings with numbers}
### Limitations
{2-3 honest limitations}
---
## 5) Why it matters — 对我研究的启发
{2-4 numbered insights connecting to reader's research}
## 6) Actionable next step
- [ ] {Follow-up item 1}
- [ ] {Follow-up item 2}
- [ ] {Follow-up item 3}
## 7) 评分解释
**{N}/5（{BREAKDOWN}）**
{Bullet explanation for each component}Show full SKILL.md (216 more words)Show less
Key rules for markdown deep note:
- Use real numbers — never write "XX" or placeholders
- Figures must have URLs from arxiv HTML version when available
- Tables use pipe format — compatible with GitHub/Obsidian
- Chinese + English mixed — technical terms in English, analysis in Chinese
- Scoring breakdown must be explicit — show the math
---## Step 5 — Output Chat Summary
After writing the markdown file, output a brief summary in chat:
📝 Deep Note 完成 | {ALIAS}
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
📌 **{TITLE}**
👤 {AUTHORS} | 📅 {YEAR} | ⭐ {RATING}
🔗 https://arxiv.org/abs/{ARXIV_ID}
💡 **核心发现：** {WHY_READ one sentence}
📊 **关键数据：** {BEST_RESULT metric + number}
✨ **启发：** {TOP_INSIGHT one sentence}
💾 笔记 → {OUTPUT_PATH}Step 6 — (Optional) Generate HTML
Only if user requests HTML (--html / 生成 HTML / 生成网页):
- Load template from
{SKILL_DIR}/../../templates/paper-note.html - Fill all
{{PLACEHOLDER}}tags using extracted content - Save to output directory as
{ARXIV_ID}.html - Report the saved path
See the root SKILL.md for the full placeholder mapping table.
Error Handling
| Error | Action |
|---|---|
| arXiv HTML unavailable | Use abstract page only; note [HTML unavailable] |
| Very long paper (>50 pages) | Focus on Abstract, Intro, Method, Results tables, Conclusion |
| No figures found in HTML | Skip Figure section, note [No figures extracted] |
| Config file missing | Use defaults; suggest "更新我的研究画像" |
Token Efficiency Notes
- Markdown deep note costs ~2-3K output tokens (vs ~8-10K for HTML template filling)
- Markdown files are git-friendly: diff, merge, grep all work naturally
- Reading list tables can directly link to markdown notes via relative paths
- HTML generation is now opt-in, not default — saves tokens on every paper read
© AlphaLab-USTC, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in skills/paper-reader of AlphaLab-USTC/ResearchClaw.
Open the folder on GitHubat commit 9d64c4b
Compare with similar skills
Paper Reader next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Paper Reader this skillAlphaLab-USTC/ResearchClaw | 134 | — | ~1.6k | Automated safety check: Pass | MIT | |
| Nature Readerjing1312/nature-figure-skill | 167 | — | ~2.8k | Automated safety check: Pass | MIT | |
| Paper To HTMLysyecust/lecture-to-notes | 270 | — | ~1.3k | Automated safety check: Pass | Custom licence | |
| Paper Interpretationdigoal/blog | 8.6k | — | ~1.5k | Automated safety check: Pass | GPL-2.0 | |
| Write Paperfrenzymath/Danus | 475 | — | ~18k | Automated safety check: Pass | Apache-2.0 | |
| Fulltext RetrievalAperivue/medsci-skills | 329 | — | ~1.9k | Automated safety check: Pass | MIT |
Similar skills
Paper To HTML
ysyecust/lecture-to-notes
Generate a self-contained, beautifully styled HTML analysis of an academic paper.
Paper Interpretation
digoal/blog
从论文 PDF 文件或论文 PDF URL 生成通俗易懂、图文并茂、带批判性评估的中文 Markdown 解读，并保存到当前项目的 markdown 目录。Use when the user asks to interpret,精读,解读,summarize,explain,analyze, or write an article from an academic paper PDF…
Write Paper
frenzymath/Danus
Turn a project's verified fact graph into a publishable LaTeX paper in a configurable house style — a standalone amsart .tex with a real bibliography, compiled to PDF.
Fulltext Retrieval
Aperivue/medsci-skills
A skill your agent uses when you need full-text PDFs for a list of DOIs, such as a meta-analysis screening set.
Read arXiv Paper
karpathy/nanochat
Fetches the TeX source of an arXiv paper from its URL, reads it and writes a markdown summary tied to the nanochat project.
More from AlphaLab-USTC/ResearchClaw
Idea Generator
AlphaLab-USTC/ResearchClaw
Analyze connections across read papers and generate actionable research ideas.
Paper Scout
AlphaLab-USTC/ResearchClaw
Discover and recommend latest arXiv papers matching user research interests.
Paper Writer
AlphaLab-USTC/ResearchClaw
Help write research papers: outline, draft sections, auto-review, and rebuttal.
Reading List
AlphaLab-USTC/ResearchClaw
Manage a personal reading list with kanban-style statuses and HTML dashboard.
Research Claw
AlphaLab-USTC/ResearchClaw
Complete AI research assistant with 6 core capabilities: paper discovery, deep reading notes, reading list management, research taste learning, idea generation, and paper writing.
Research Profile
AlphaLab-USTC/ResearchClaw
Maintain and visualize the user research preference profile.
Related
Categories
Questions about Paper Reader
What does Paper Reader do?
Deep-read an arXiv paper and generate structured Deep Note reading notes. Paper Reader is an agent skill from AlphaLab-USTC/ResearchClaw. Deep-read an arXiv paper and generate structured Deep Note reading notes.
When should I use Paper Reader?
Paper Reader fits situations like: tasks that involve Academic paper search; tasks that involve HTML artifacts; tasks that involve LLM cost and token optimization.
How do I install Paper Reader in Claude Code?
Run `npx skills add AlphaLab-USTC/ResearchClaw --skill paper-reader -a claude-code`. Or copy the skill folder (skills/paper-reader in AlphaLab-USTC/ResearchClaw) into .claude/skills/paper-reader in your project. Claude Code loads it when a task matches its description.
How do I install Paper Reader in Codex?
Run `npx skills add AlphaLab-USTC/ResearchClaw --skill paper-reader -a codex`. Or copy the skill folder (skills/paper-reader in AlphaLab-USTC/ResearchClaw) into .agents/skills/paper-reader in your project. Codex loads it when a task matches its description.
Can I use Paper Reader in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add AlphaLab-USTC/ResearchClaw --skill paper-reader -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/paper-reader, .gemini/skills/paper-reader, .github/skills/paper-reader and .opencode/skills/paper-reader in your project.
What does Paper Reader need to run?
SKILL.md names no scripts, command-line tools or credentials: Paper Reader is instructions for the agent only.
Does Paper Reader access the network?
SKILL.md names 1 domain. In commands or code: arxiv.org; the agent is likely to contact it when it follows the instructions. This is read from the text; nothing was executed.
Is Paper Reader safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Paper Reader use?
Paper Reader is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Paper Reader use?
About 1.6k tokens (SKILL.md is roughly 6.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Paper Reader?
Skills that share tags, products or a category with Paper Reader: Nature Reader (jing1312/nature-figure-skill, 167 stars), Paper To HTML (ysyecust/lecture-to-notes, 270 stars), Paper Interpretation (digoal/blog, 8.6k stars) and Write Paper (frenzymath/Danus, 475 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Paper Reader?
AlphaLab-USTC (a GitHub user) maintains it in AlphaLab-USTC/ResearchClaw, which has 134 GitHub stars. The repository holds 7 skills in this directory. The repository was last updated on April 7, 2026.
Source: AlphaLab-USTC/ResearchClaw on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.