AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

论文讲解助手

skill-laborany-laborany-paper-explainer · by laborany

|

No reviews yet
0 installs
26 views
0.0% view→install

Install

$ agentstack add skill-laborany-laborany-paper-explainer

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-laborany-laborany-paper-explainer)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
4mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of 论文讲解助手? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

论文讲解助手

将复杂学术论文转化为结构化、易理解的知识文档。

工作流程

PDF输入 → 解析提取 → 深度分析 → HTML输出

Step 1: PDF解析

运行 scripts/parse_pdf.py 提取原始内容:

python scripts/parse_pdf.py  -o parsed.json --image-dir ./images

输出结构:

{
  "pages": [{"page_num": 1, "text": "...", "tables": [...]}],
  "images": [{"page_num": 1, "image_index": 1, "path": "..."}]
}

Step 2: 内容分析

阅读解析结果,提取以下信息:

| 字段 | 来源 | 说明 | |------|------|------| | title | 首页顶部 | 论文标题 | | authors | 标题下方 | 作者列表 | | affiliations | 脚注/作者下 | 机构信息 | | motivation | Abstract + Intro | 研究动机与问题 | | method | Method章节 | 核心方法详解 | | experiments | Experiments章节 | 实验设置与结果 |

分析要点 (详见 [references/analysisguide.md](references/analysisguide.md)):

  • 动机: 回答What/Why/Gap三问
  • 方法: 分层讲解(直觉→架构→细节→数学)
  • 公式: 提供符号表+直觉解释
  • 实验: 批判性分析基线公平性

Step 2.5: 图片智能分类与嵌入

对提取的图片进行分类,识别其用途:

| 类型 | 特征 | 嵌入位置 | |------|------|----------| | 框架图 | 展示整体架构/流程,通常较大,含模块和箭头 | method 开头 | | 模块细节图 | 展示单个组件内部结构 | method 对应段落 | | 实验曲线 | 折线图/柱状图,含坐标轴和图例 | experiments 对应分析处 | | 可视化结果 | 热力图/注意力图/生成样本 | experiments 定性分析处 | | 示意图 | 概念解释/对比图 | motivation 或 method | | 其他 | Logo/装饰/无关图片 | 仅放附录或忽略 |

分类方法:

  1. 查看图片尺寸: 框架图通常宽度 > 高度,且尺寸较大
  2. 查看所在页码: 第1-2页多为示意图,Method章节多为架构图
  3. 结合论文正文中的 "Figure X" 引用,匹配图片与描述
  4. 分析图片内容: 含箭头/模块框的是架构图,含坐标轴的是实验图

嵌入策略:

  • 框架图: 在 method 开头用 `` 标签嵌入,配详细说明
  • 实验图: 在 experiments 对应结论处嵌入,解释图中趋势
  • 其他关键图: 根据论文引用位置,嵌入对应段落

Step 3: 生成HTML

构造分析结果JSON:

{
  "title": "论文标题",
  "authors": "作者1, 作者2",
  "affiliations": "机构1; 机构2",
  "motivation": "HTML格式的动机分析",
  "method": "HTML格式的方法讲解,支持$LaTeX$公式",
  "experiments": "HTML格式的实验分析",
  "images": [...],
  "embedded_images": {
    "motivation": [{"index": 0, "caption": "图1说明", "position": "after_intro"}],
    "method": [{"index": 1, "caption": "框架图说明", "position": "start"}],
    "experiments": [{"index": 2, "caption": "实验结果图", "position": "inline"}]
  }
}

embedded_images 字段说明:

  • index: 对应 images 数组中的索引
  • caption: 图片说明文字
  • position: 嵌入位置 (start/inline/end)

运行生成脚本:

python scripts/generate_html.py analysis.json -o 论文讲解.html

输出规范

LaTeX公式

  • 行内公式: $E=mc^2$
  • 独立公式: $$\sum_{i=1}^n x_i$$

内容格式

子标题
段落文本,支持代码和强调
列表项

图片处理

智能嵌入 (推荐):

  • 框架图/架构图: 嵌入 method 区域开头,配详细图注
  • 实验结果图: 嵌入 experiments 对应分析段落
  • 概念示意图: 嵌入 motivation 帮助理解问题

嵌入语法:


  
  图1: 模型整体架构。输入经过编码器...

附录处理:

  • 所有图片仍在"图表说明"区域保留完整列表
  • 嵌入的图片会同时出现在正文和附录

依赖

pip install pdfplumber PyMuPDF

快速示例

# 1. 解析PDF
python scripts/parse_pdf.py attention.pdf -o parsed.json

# 2. 分析内容 (Claude完成)
# 生成 analysis.json

# 3. 生成HTML
python scripts/generate_html.py analysis.json -o attention_explained.html

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.