ClawHub

url2pdf-mk

v1.1.1

EN: Webpage to PDF + Markdown converter. Converts any URL (especially WeChat articles) into offline-readable PDF and Markdown files with complete images, pre...

⭐ 0· 151·0 current·0 all-time
byQKZ@qinkaizhou

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for qinkaizhou/url2pdf-mk.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "url2pdf-mk" (qinkaizhou/url2pdf-mk) from ClawHub.
Skill page: https://clawhub.ai/qinkaizhou/url2pdf-mk
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install url2pdf-mk

ClawHub CLI

Package manager switcher

npx clawhub@latest install url2pdf-mk
Security Scan
VirusTotalVirusTotal
Pending
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description (webpage → PDF/Markdown, WeChat articles) match the provided scripts: HTTP scraping, browser (CDP) scraping, batch/xlsx handling, Markdown/PDF generation. Requiring a Chrome/Chromium installation and Python libraries is appropriate for the described functionality.
ℹ
Instruction Scope
SKILL.md and scripts instruct launching/controlling a local browser via CDP, reading an input xlsx, creating date folders on the desktop, and writing output files and temporary proxy state files. The skill explicitly documents that browser mode reuses the user's Chrome profile (access to cookies/login sessions) and offers an --isolated flag. These actions are within the scope of the stated purpose but have privacy implications the user should be aware of.
✓
Install Mechanism
No remote download/install spec is present; this is an instruction/code bundle with local Python scripts. No external binary downloads or strange URLs are used. Dependencies are standard Python packages (requests, openpyxl, websockets, etc.) and the browser binary (Chrome/Chromium).
ℹ
Credentials
The skill declares no required environment variables or credentials. However, by default it may reuse the user's Chrome profile directory (platform-specific profile paths) and will read/write temp state files (cdp-proxy-<user>) and output to Desktop—access that is proportionate to scraping but can expose cookies, session tokens, and local profile data. The SKILL.md warns about this and offers an isolated mode.
✓
Persistence & Privilege
always:false and no special platform-wide modifications are requested. The CDP proxy may spawn a local helper process and write proxy state files in the temp directory while running, but nothing indicates it persists beyond normal runtime files or modifies other skills/configs.
Assessment
This skill appears to do what it says, but it controls a real browser and by default reuses your Chrome/Chromium profile. That means if you run it in default (non-isolated) mode the script can access cookies, logged-in sessions, and other profile data. The SKILL.md already warns about this; before using the skill: (1) prefer --isolated mode if you don't want the tool to access your login sessions, (2) run it on non-sensitive accounts or inside a VM/container if you need extra isolation, (3) check the xlsx input file is from a trusted source (the script will read it and access the listed URLs), (4) be aware the tool opens a local CDP proxy port and writes temp files (cdp-proxy-<user>), and (5) review/verify the included scripts and Python dependencies before running. If you want to avoid any possibility of the agent autonomously launching the browser against your profile, disable model/skill invocation or run the tool manually with explicit command-line flags.

Like a lobster shell, security has layers — review code before you run it.

latestvk97fzm53m7cm83wmaqg7cwxaz184t633
151downloads
0stars
10versions
Updated 2w ago
v1.1.1
MIT-0
QKZ@qinkaizhou
latest
Download

url2pdf-mk — Webpage to PDF & Markdown Converter

One-click webpage to PDF + Markdown converter. Transforms WeChat articles and web pages into offline-readable documents with full images, original layout and styles preserved.

一键将网页（尤其是微信文章）转换为可离线阅读的 PDF 和 Markdown 文档，完整保留排版、图片、样式。

📋 概述 | Overview

url2pdf-mk is a professional webpage content scraping and conversion tool, designed specifically for WeChat public account articles and regular web pages. It can:

  • ✅ Convert any webpage into PDF and Markdown dual-format output / 将任意网页转换为 PDF 和 Markdown 双格式输出
  • ✅ Preserve complete images, layout and styles / 完整保留网页中的图片、排版、样式
  • ✅ Support both single-page and batch scraping (via xlsx file) / 支持单篇抓取和批量抓取
  • ✅ Auto-create date-named folders on desktop / 自动按日期在桌面创建归档目录
  • ✅ Smart detection of WeChat article titles and publish dates / 智能识别微信文章标题和发布日期

⚙️ 工作机制 | How It Works

核心技术架构 | Core Architecture

┌─────────────────────────────────────────────────────────────────┐
│                    url2pdf-mk Workflow                           │
│                        工作流程                                   │
├─────────────────────────────────────────────────────────────────┤
│                                                                 │
│  Input Sources / 输入源                                          │
│  ├── Single URL → scrape.py (Single Page)                       │
│  │   └── 单个 URL → 单篇抓取                                     │
│  ├── Multiple URLs → batch_scrape.py (Browser Batch)            │
│  │   └── 多个 URL → 批量浏览器版                                  │
│  └── xlsx file → Smart Routing (Browser/HTTP)                  │
│      └── xlsx 文件 → 智能路由                                    │
│                                                                 │
│  Scraping Engines / 抓取引擎                                      │
│  ├── Browser (CDP) ──→ Chrome DevTools Protocol                 │
│  │   ├── Launch Chrome/Chromium browser                          │
│  │   │   └── 启动浏览器                                           │
│  │   ├── Control browser via CDP protocol                        │
│  │   │   └── CDP 协议控制                                         │
│  │   ├── Render full page (including JavaScript)                │
│  │   │   └── 渲染完整页面                                         │
│  │   └── Extract DOM + screenshot to PDF                        │
│  │       └── 提取 DOM + 截图生成 PDF                               │
│  │                                                              │
│  └── HTTP (No Browser) ──→ requests library                      │
│      ├── Send HTTP requests to get HTML                         │
│      │   └── HTTP 请求获取 HTML                                   │
│      ├── Parse static HTML structure                            │
│      │   └── 解析静态 HTML                                        │
│      └── Markdown output only (no PDF)                          │
│          └── 仅输出 Markdown                                     │
│                                                                 │
│  Output Processing / 输出处理                                     │
│  ├── HTML → Markdown conversion (preserve image links)          │
│  │   └── HTML → Markdown 转换                                    │
│  ├── HTML → PDF generation (reportlab + full styles)            │
│  │   └── HTML → PDF 生成                                          │
│  └── Filename: {date}_{title}.{md/pdf}                          │
│      └── 文件命名：{发布日期}_{标题}                               │
│                                                                 │
└─────────────────────────────────────────────────────────────────┘

工作流程详解 | Workflow Details

1. Single Page Mode / 单篇抓取模式 (scrape.py)

URL → Launch browser → Load page → Extract content → Generate PDF + Markdown → Save to date folder
URL → 启动浏览器 → 加载页面 → 提取内容 → 生成 PDF + Markdown → 保存到日期目录

2. Browser Batch Mode / 批量浏览器版 (batch_scrape.py)

xlsx file → Read URL list → Reuse browser instance → Scrape each → Batch output
xlsx 文件 → 读取 URL 列表 → 复用浏览器实例 → 逐个抓取 → 批量输出
  • Advantage / 优势: Browser instance starts only once, high efficiency / 浏览器实例只启动一次，效率高
  • Prerequisite / 前提: Chrome/Chromium installed / 系统已安装 Chrome/Chromium

3. HTTP Batch Mode / 批量 HTTP 版 (batch_http.py)

xlsx file → Read URL list → Send HTTP requests → Parse HTML → Batch output Markdown
xlsx 文件 → 读取 URL 列表 → 发送 HTTP 请求 → 解析 HTML → 批量输出 Markdown
  • Advantage / 优势: No browser needed, low resource usage / 无需浏览器，资源占用低
  • Limitation / 限制: Cannot handle JavaScript-rendered content, no PDF output / 无法处理 JS 内容，无 PDF 输出

智能路由逻辑 | Smart Routing Logic

main.py serves as the unified entry point, automatically selecting the optimal approach based on input type:

Input Type / 输入类型Chrome Available / Chrome 可用Chrome Unavailable / Chrome 不可用
Single URL / 1 个 URLBrowser mode (PDF+MD) / 浏览器版HTTP mode (MD only) / HTTP 版
Multiple URLs / ≥2 个 URLBrowser batch (PDF+MD) / 浏览器版批量HTTP batch (MD only) / HTTP 版批量
xlsx file / xlsx 文件Browser batch (PDF+MD) / 浏览器版批量HTTP batch (MD only) / HTTP 版批量

CDP (Chrome DevTools Protocol) / CDP 协议说明

CDP is Chrome's debugging protocol. This tool uses it to:

  1. Launch/Connect browser / 启动/连接浏览器: Connect to existing browser or launch new instance / 支持连接已有浏览器或启动新实例
  2. Page control / 页面控制: Navigate, wait for load, execute JavaScript / 导航、等待加载、执行 JavaScript
  3. Content extraction / 内容提取: Get complete DOM structure and computed styles / 获取完整的 DOM 结构和计算样式
  4. PDF generation / PDF 生成: Use browser's native PDF print function / 调用浏览器原生 PDF 打印功能

Security measures / 安全措施:

  • CDP ports bound to 127.0.0.1 only, external network cannot access / CDP 端口仅绑定本机，外网无法访问
  • WebSocket connections require Token verification / WebSocket 连接需要 Token 验证
  • --isolated mode uses independent temporary Profile, no access to user data / --isolated 模式使用临时 Profile，不访问用户数据

🖥️ 适配环境 | Supported Environments

操作系统支持 | Operating System Support

Operating System / 操作系统Status / 支持状态Note / 说明
Windows 10/11✅ Full Support / 完全支持Chrome or Edge recommended / 推荐 Chrome 或 Edge
macOS 10.15+✅ Full Support / 完全支持Chrome, Chromium, Edge supported / 支持 Chrome、Chromium、Edge
Linux✅ Full Support / 完全支持Chrome or Chromium required / 需安装 Chrome 或 Chromium
Windows 7/8⚠️ Limited / 有限支持Upgrade recommended for best experience / 建议升级以获得最佳体验

浏览器支持 | Browser Support

Browser / 浏览器Status / 支持状态Version / 版本要求
Google Chrome✅ Recommended / 推荐80+
Microsoft Edge (Chromium)✅ Supported / 支持80+
Chromium✅ Supported / 支持80+
Others / 其他浏览器❌ Not supported / 不支持Chromium-based only / 仅支持 Chromium 内核

Python 环境要求 | Python Environment Requirements

Component / 组件Version / 版本要求Purpose / 用途
Python3.7+Runtime environment / 运行时环境
pipLatest / 最新版Package manager / 包管理器

📦 基础组件依赖 | Dependencies

Required Dependencies / 必需依赖 (must install before running)

pip install websockets openpyxl requests reportlab
Package / 依赖包Version / 版本要求Purpose / 用途
websockets10.0+CDP WebSocket communication / CDP WebSocket 通信
openpyxl3.0+Read xlsx files (batch mode) / 读取 xlsx 文件（批量模式）
requests2.25+HTTP requests (HTTP mode) / HTTP 请求（HTTP 版抓取）
reportlab3.6+PDF document generation / PDF 文档生成

Optional Dependencies / 可选依赖

Package / 依赖包Purpose / 用途Note / 说明
PillowImage processing / 图片处理PDF image optimization (auto-installed) / PDF 图片优化（自动安装）
beautifulsoup4HTML parsing / HTML 解析Enhanced content extraction (built-in) / 增强内容提取（已内置）

System Requirements / 系统组件要求

Component / 组件Requirement / 说明
Chrome/Chromium/EdgeRequired for browser mode / 浏览器版抓取必需
Network / 网络连接Required to access target webpages / 访问目标网页必需
Disk space / 磁盘空间100MB+ recommended / 建议预留 100MB+

快速环境检查 | Quick Environment Check

Run the following commands to check if the environment is ready:

# Check Python version
python3 --version  # Should show 3.7+

# 检查依赖是否已安装
python3 -c "import websockets, openpyxl, requests, reportlab; print('✅ All dependencies installed')"

# Check if Chrome is available
# Windows
where chrome
# macOS
which "Google Chrome"
# Linux
which google-chrome || which chromium-browser

🔒 安全说明（必读）| Security

Mode / 模式Command / 命令ProfileUse Case / 适用场景
Isolated (Recommended) / 隔离（推荐）main.py --isolated <URL>Temporary isolated Profile, no Cookie / 临时隔离 Profile，无 CookiePublic content / 公开内容
Default / 默认main.py <URL>Reuse Chrome real Profile / 复用 Chrome 真实 ProfileWeChat articles requiring login / 需登录态的微信文章

风险与缓解 | Risks & Mitigations

Risk / 风险项Description / 说明Mitigation / 缓解措施
Profile reuse / Profile 复用Can access Chrome cookies/login sessions / 可访问 Chrome Cookie/登录态✅ Always use --isolated for public content / 公开内容一律用 --isolated
CDP remote debugging / CDP 远程调试Enables Chrome DevTools protocol / 开启 Chrome DevTools 协议Only listens on 127.0.0.1:9222~9232 (localhost) / 仅监听本机
CDP ProxyCreates local proxy daemon / 创建本地代理守护进程Only bound to 127.0.0.1, external access blocked / 仅绑定本机；WS 需 Token 验证
Browser launch / 浏览器启动May reuse existing browser window / 可能复用已有浏览器窗口--isolated always launches independent instance / 始终启动独立实例
sys.pathOnly adds skill directory, no external path access / 仅添加本技能目录✅ Already securely isolated / 已安全隔离
pip dependencies / pip 依赖No longer auto-install at runtime / 不再运行时自动安装✅ Pre-installed + error prompt / 预装 + 报错提示

推荐规则 | Recommended Rules

⚠️ Default mode reuses your real Chrome profile — it can read cookies and logged-in sessions. Always prefer --isolated for public content to avoid exposing your browser session.

  • Public content / 公开内容 → --isolated (No login required, uses temporary isolated profile / 无需登录，使用临时隔离 Profile)
  • Content requiring login / 需登录的内容 → Default mode (Ensure browser is logged in / 确保浏览器已登录目标网站；或在专用/沙箱环境中运行)
  • Safest option / 最安全方案 → HTTP batch mode (batch_http.py) — no browser, no profile access / 无需浏览器，不访问 Profile

⚠️ 重要安全警告 | Important Security Warnings

⭐ For maximum safety, prefer --isolated mode or HTTP batch mode (batch_http.py) so the tool never touches your real browser profile or cookies.

  1. Cookie and Login Access / Cookie 和登录态访问: Default mode connects to your existing Chrome instance, allowing access to browser cookies and logged-in sessions. Only use default mode when you trust the code and need to scrape pages requiring login.

  2. CDP Proxy — Local Port & State Files / CDP 代理 — 本地端口与状态文件: The CDP proxy daemon opens a local listening port (default 9223) and writes PID/state files to your temp directory (named cdp-proxy-<username>). After each task, manually check and delete these temp files if auto-cleanup failed.

  3. Temporary Files / 临时文件: Scripts create temporary files (xlsx, proxy state, pid files) in system temp directory. These are auto-cleaned after task completion. For extra security, manually check and delete relevant files in temp directory after tasks.

  4. CDP Port Security / CDP 端口安全: CDP debugging ports are bound to 127.0.0.1 only, external network cannot access. WebSocket connections require Token verification.

  5. Don't Run as Root / 不要以 root 运行: Run with normal user privileges recommended / 建议以普通用户权限运行此工具。

  6. VM / Sandbox / 虚拟机 / 沙箱: If you must run browser mode with your real profile, do so inside a VM or disposable environment to limit exposure.

  7. Extra Assurance / 如需额外保证: Modify browser_launcher.py to force isolated profile only.

用法 | Usage

统一入口（推荐）| Unified Entry (Recommended)

# Single page / 单篇
python3 scripts/main.py "https://mp.weixin.qq.com/s?__biz=..."

# Batch (multiple URLs) / 批量（多个 URL）
python3 scripts/main.py "https://url1" "https://url2" "https://url3"

# Batch (xlsx file) / 批量（xlsx 文件）
python3 scripts/main.py /path/to/urls.xlsx

# Isolated mode (no login) / 隔离模式（不使用登录态）
python3 scripts/main.py --isolated "https://..."
python3 scripts/main.py --isolated /path/to/urls.xlsx

Smart Routing / 智能路由

Input / 输入Processing / 处理方式
Single URL / 1 个 URLSingle-page full scraping (PDF + Markdown) / 单篇完整抓取
Multiple URLs / ≥2 个 URLBatch scraping: Browser mode if Chrome available, otherwise HTTP mode / Chrome 可用 → 浏览器版，否则 → HTTP 版
xlsx file / xlsx 文件Auto-routed based on URL count / 根据 URL 数量自动路由

xlsx 格式 | xlsx Format

Column / 列Content / 内容
B (2nd) / B（第2列）Article title / 文章标题
C (3rd) / C（第3列）Publish date / 发布日期
F (6th) / F（第6列）Article URL / 文章 URL

输出 | Output

~/Desktop/{current date}/
├── 2025-04-03_article-title.md
└── 2025-04-03_article-title.pdf

脚本说明 | Scripts

Script / 脚本Purpose / 用途
main.pyUnified entry, smart routing / 统一入口，智能路由
scrape.pySingle-page scraping, PDF + Markdown / 单篇抓取，PDF + Markdown
batch_scrape.pyBrowser batch (requires Chrome) / 批量浏览器版（需 Chrome）
batch_http.pyHTTP batch (no browser, MD only) / 批量 HTTP 版（无浏览器，仅 Markdown）

已知限制 | Known Limitations

  • Video/Audio / 视频/音频: PDF keeps cover, Markdown keeps links / PDF 保留封面，Markdown 保留链接
  • WeChat Mini Programs / 微信小程序: Static content only / 仅提取静态内容
  • Anti-crawl limits / 反爬限制: HTTP mode may fail, switch to browser mode / HTTP 版可能失败，改用浏览器版

更新日志 | Changelog

v1.1.1（2026-04-14）— 安全描述增强

  • SKILL.md description metadata 新增安全警告：
    • 明确告知浏览器默认模式会复用 Chrome Profile，可访问 Cookie/登录态
    • 提示 CDP 代理会打开本地端口并写入临时状态文件
    • 推荐公开内容优先使用 --isolated
  • 推荐规则强化：
    • 新增最安全方案：HTTP 批量模式（batch_http.py），无需浏览器，不访问 Profile
    • 所有推荐规则前加 ⭐ 标识
  • ⚠️ 安全警告章节扩充：
    • 新增 CDP 代理本地端口与 cdp-proxy-<username> 状态文件提示
    • 新增 VM / 沙箱运行建议
    • 临时文件手动清理建议

v1.1.0（2026-04-14）— English-first bilingual formatting

  • 🌐 全面中英双语调整 / English-first throughout:
    • All section titles adjusted to "English | 中文" format
    • All tables adjusted to English first, Chinese second
    • All descriptions adjusted to English first, Chinese second
    • Architecture diagram converted to bilingual layout
    • Security warnings, usage instructions, and known limitations all bilingual
    • description metadata: EN description placed before Chinese

v1.0.9（2026-04-14）— 文档增强

SKILL.md 文档大幅更新：

  • 📋 新增「概述」章节 / Overview section: 简要说明工具核心功能定位
  • ⚙️ 新增「工作机制」/ How It Works 章节:
    • 核心技术架构流程图
    • 工作流程详解（单篇抓取、批量浏览器版、批量 HTTP 版）
    • 智能路由逻辑表格
    • CDP 协议技术说明
  • 🖥️ 新增「适配环境」/ Supported Environments 章节:
    • 操作系统支持表（Windows/macOS/Linux）
    • 浏览器支持表（Chrome/Edge/Chromium 版本要求）
    • Python 环境要求
  • 📦 新增「基础组件依赖」/ Dependencies 章节:
    • 必需依赖包及版本要求（websockets、openpyxl、requests、reportlab）
    • 可选依赖说明
    • 系统组件要求
    • 快速环境检查命令
    • 便于 SkillHub 平台海外用户理解工具功能

文档结构优化：

  • 将「环境」章节整合到「基础组件依赖」中，避免重复
  • 统一表格格式，提升可读性

v1.0.8（2026-04-13）— 安全增强

  • SKILL.md：新增**⚠️ 重要安全警告**章节，包含：
    • Cookie 和登录态访问警告
    • 临时文件安全说明
    • CDP 端口安全说明
    • 不要以 root 运行的建议
    • VM/disposable 环境运行建议
  • main.py：临时文件清理添加日志输出，清理成功/失败都会提示

v1.0.7（2026-04-13）— 文件名修复

  • 批量抓取时，文件名（PDF + Markdown）现在使用从网页提取的真实标题和日期
  • 修复了之前使用 xlsx 默认标题（如"文章1"）的问题
  • batch_scrape.py：在提取网页内容后，用网页获取的真实标题覆盖 xlsx 中的标题

v1.0.5（2026-04-13）— 安全修复

  • 移除 batch_http.py 中的硬编码 xlsx 默认路径
  • 所有脚本：pip 依赖从运行时 auto-install 改为预装检查 + 报错提示（安全加固）
  • batch_http.py 新增系统目录路径安全校验
  • SKILL.md：调整安全说明，--isolated 标记为公开内容首选

v1.0.4（2026-04-13）

  • 新增 --isolated 隔离模式参数
  • 移除硬编码路径，必须提供 xlsx 路径
  • 新增 --output 可指定输出目录
  • 强化安全文档

v1.0.1（2026-04-12）

  • 新增批量抓取脚本
  • 修复多项 bug

Comments

Loading comments...