← Back to Skills Marketplace
230
Downloads
0
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install paper-reading
Description
Deep-dive analysis of academic papers, with optional architecture diagrams of the paper's method. Use this skill whenever the user shares a research paper, a...
Usage Guidance
This skill appears coherent with its stated purpose and requests no credentials, installs, or risky downloads. Before enabling it, consider: (1) Automatic triggering — the skill's instructions explicitly tell the agent to run when 'paper content is present' (even without explicit invocation); if you do not want automatic analysis whenever a PDF or paper text appears, disable/autoinvoke controls or avoid posting such content in contexts where the agent is active. (2) Persistence — reports and diagrams are written to /mnt/user-data/outputs or a workspace outputs/ directory; uploaded or pasted papers (including unpublished manuscripts) and extracted text will be persisted there. (3) External resources — the diagram HTML references Google Fonts (and rendering HTML/SVG may cause the client to load remote resources); expect normal web requests for rendering. If these behaviors are acceptable for your workflow (automatic analysis + local file outputs + benign web fetches), the skill is internally consistent and reasonable to use. If you are concerned about auto-triggering or storing sensitive papers, do not enable it or restrict its invocation.
Capability Analysis
Type: OpenClaw Skill
Name: paper-reading
Version: 1.1.0
The skill bundle is a comprehensive tool designed for academic paper analysis and architecture diagram generation. It utilizes standard capabilities such as 'web_fetch' for retrieving arXiv content and Playwright for rendering HTML diagrams into PNG format. The instructions in SKILL.md and the supporting reference files (e.g., diagram-design.md, output-cn.md) are highly detailed and strictly focused on the stated task of producing rigorous, peer-review-style reports. No evidence of malicious intent, data exfiltration, or harmful prompt injection was found; all high-risk actions (file writing and browser automation) are logically tied to the skill's functional requirements.
Capability Tags
Capability Assessment
Purpose & Capability
The name/description (paper deep reading + diagram drawing) align with the instructions and assets: SKILL.md explains fetching arXiv HTML, reading PDFs, producing a Markdown report, and optionally rendering diagrams using the provided HTML/SVG template. No unrelated binaries, env vars, or config paths are requested.
Instruction Scope
Instructions direct the agent to fetch arXiv HTML (via web_fetch), use pdf-reading tools for PDFs, extract all sections including appendices, and always write a report file to /mnt/user-data/outputs (with a workspace fallback). This is consistent with the stated purpose, but two operational notes deserve attention: (1) SKILL.md says 'Don't wait for explicit skill invocation — if paper content is present, use it', meaning the agent may autonomously trigger whenever paper-like content appears; (2) the skill mandates writing files to disk (reports and diagrams), so uploaded/unpublished paper contents and extracted text will be persisted to the agent workspace unless the platform enforces otherwise.
Install Mechanism
There is no install spec and no code files that get executed on install — this is instruction-only plus static assets (HTML templates and reference docs). That minimizes installation risk; nothing is downloaded or extracted at install time.
Credentials
The skill requires no environment variables, credentials, or config paths. It does reference external web resources (arXiv HTML, Google Fonts in the diagram template) which are appropriate for the described functionality and do not imply secret access requirements.
Persistence & Privilege
always:false and no special privileges are requested. Autonomous invocation is permitted (disable-model-invocation:false) which is the platform default; combined with the SKILL.md trigger rules, the agent may auto-run when encountering paper content. The skill writes files into /mnt/user-data/outputs (or outputs/ fallback) — normal for this purpose but users should expect persisted artifacts in their workspace.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install paper-reading - After installation, invoke the skill by name or use
/paper-reading - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.0
**paper-reading v1.1.0 Changelog**
- Added optional architecture diagram output: upon request or flag (`--draw-structure`), the skill now generates a publication-quality architecture diagram of the paper's method.
- Introduced support for diagram-related flags, including output format (`--output`), background theme (`--bg-theme`), and separate diagram/report language (`--showcase-language`).
- Significantly expanded README documentation in both English and Chinese.
- Added reference guides for diagram design and labels, supporting structured and localized diagram creation.
- Updated analysis workflow and output guidance to always produce a structured Markdown report and diagrams as requested.
v1.0.0
**English**
> A Claude skill for deep-diving academic papers. Upload a PDF or paste paper content — Claude automatically produces a structured six-section analysis covering problem framing, technical method, experimental evidence, critical review, and synthesis. All claims are tagged by source (`[paper]` / `[inferred]` / `[external]`). Outputs in English or Chinese based on the language you write in.
---
**中文**
> 一个用于深度解读学术论文的 Claude Skill。上传 PDF 或粘贴论文内容，Claude 自动输出包含六个章节的结构化分析：研究问题、技术方案、实验验证、批判性评审与综合总结。所有论断均标注信息来源（`[paper]` / `[inferred]` / `[external]`），根据你的输入语言自动切换中英文输出。
Metadata
Frequently Asked Questions
What is Academic paper reading, summarizing, polished method diagrams drawing SKILL?
Deep-dive analysis of academic papers, with optional architecture diagrams of the paper's method. Use this skill whenever the user shares a research paper, a... It is an AI Agent Skill for Claude Code / OpenClaw, with 230 downloads so far.
How do I install Academic paper reading, summarizing, polished method diagrams drawing SKILL?
Run "/install paper-reading" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Academic paper reading, summarizing, polished method diagrams drawing SKILL free?
Yes, Academic paper reading, summarizing, polished method diagrams drawing SKILL is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Academic paper reading, summarizing, polished method diagrams drawing SKILL support?
Academic paper reading, summarizing, polished method diagrams drawing SKILL is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Academic paper reading, summarizing, polished method diagrams drawing SKILL?
It is built and maintained by Wonster (@wooonster); the current version is v1.1.0.
More Skills