Screenshot Tool

网页截图 + 文档截图工具。支持网页全页截图、PPT/Word/Excel/PDF 转高清图片。保留原始样式，300 DPI 高清输出。

Security Scan

The skill's requirements, instructions, and included scripts are consistent with a webpage and document screenshot/convert tool; no unexplained credentials, endpoints, or installs are present, though the scripts call shell commands and rely on an external agent-browser CLI so typical operational cautions apply.

Purpose & Capability

Name/description (web screenshots and document→image conversion) align with declared binaries (libreoffice, pdftoppm/ poppler, agent-browser) and the included Python scripts implement exactly those flows.

Instruction Scope

SKILL.md and scripts stick to the stated task (convert documents via LibreOffice+pdf2image and screenshot via agent-browser). However, the scripts spawn shell commands (subprocess.run with shell=True) that interpolate user-provided paths/URLs; this is expected for a CLI wrapper but can be vulnerable to shell injection if untrusted/unsanitized inputs are passed.

Install Mechanism

There is no automated install spec (instruction-only), but SKILL.md directs users to install agent-browser via npm and system packages via apt — these are reasonable for the functionality. As with any third-party CLI installed from npm, users should verify the agent-browser package source and trustworthiness before installing.

Credentials

No environment variables, credentials, or config paths are requested; requested binaries match the tool's needs and are proportionate to the described capability.

Persistence & Privilege

Skill does not request always:true or other persistent platform privileges; it is user-invocable and can be invoked autonomously (platform default), which is normal for skills.

Assessment

This skill appears to do what it claims: webpage screenshots (via agent-browser) and document→image conversion (LibreOffice → pdf2image). Before installing or running: 1) Verify the agent-browser npm package/source (npm packages can execute code and download browser dependencies). 2) Only run the scripts on files/URLs you trust — the scripts construct shell commands with user-supplied filenames/URLs and use shell=True, which can be abused if an attacker controls the input (sanitize or avoid untrusted filenames). 3) Install system packages (LibreOffice, poppler-utils) from your OS package manager to avoid unknown binaries. If you need higher assurance, review or run the Python scripts in a sandbox and/or replace subprocess shell usage with safer invocation (list-argument subprocess.run).

Screenshot Tool - 网页 & 文档截图工具

支持网页截图和文档转高清图片，保留原始样式。

功能

功能	说明
网页截图	使用 headless 浏览器截图，支持整页、单页
文档转图	PPT/Word/Excel/PDF 转 300 DPI 高清图片
高清输出	4000×2250 像素，适合打印和展示

依赖安装

必需依赖

依赖	用途	安装命令
agent-browser	网页截图	npm install -g agent-browser && agent-browser install
LibreOffice	文档转 PDF	sudo apt-get install -y libreoffice-impress libreoffice-writer libreoffice-calc
poppler-utils	PDF 处理	sudo apt-get install -y poppler-utils
Python 库	PDF 转图片	pip3 install pdf2image pillow

安装步骤

# 1. 安装 agent-browser（网页截图必需）
npm install -g agent-browser
agent-browser install
agent-browser install --with-deps  # 如需要系统依赖

# 2. 安装 LibreOffice（文档转换必需）
sudo apt-get install -y libreoffice-impress libreoffice-writer libreoffice-calc

# 3. 安装 poppler-utils（PDF 处理必需）
sudo apt-get install -y poppler-utils

# 4. 安装 Python 依赖
pip3 install pdf2image pillow

# 5. 安装中文字体（可选，用于中文文档）
sudo apt-get install -y fonts-wqy-zenhei fonts-wqy-microhei fonts-noto-cjk

验证安装

# 验证 agent-browser
agent-browser --version

# 验证 LibreOffice
libreoffice --version

# 验证 poppler
which pdftoppm pdfinfo

使用方法

1. 网页截图

# 截图单个网页
python3 skills/screenshot-tool/scripts/web_screenshot.py --url "https://example.com" --output page.png

# 截图并滚动（长页面）
python3 skills/screenshot-tool/scripts/web_screenshot.py --url "https://example.com" --full-page --output page.png

2. 文档转图片

# PPT/Word/Excel/PDF 转图片
python3 skills/screenshot-tool/scripts/doc_screenshot.py --input file.pptx --output-dir ./images

# 指定 DPI（默认 300）
python3 skills/screenshot-tool/scripts/doc_screenshot.py --input file.pdf --dpi 200 --output-dir ./images

3. 使用 agent-browser 截图

# 打开网页
agent-browser open "https://example.com" --timeout 60000

# 截图
agent-browser screenshot output.png --full

# 关闭浏览器
agent-browser close

支持的格式

文档格式

格式	扩展名	状态
PowerPoint	.pptx, .ppt	✅ 支持
Word	.docx, .doc	✅ 支持
Excel	.xlsx, .xls	✅ 支持
PDF	.pdf	✅ 支持
OpenDocument	.odp, .odt, .ods	✅ 支持

网页截图

方式	说明	依赖
agent-browser	使用 headless Chrome	agent-browser
OpenClaw browser	内置浏览器工具	OpenClaw 内置

输出规格

参数	默认值	说明
DPI	300	分辨率
格式	PNG	图片格式
尺寸	4000×2250	16:9 比例

示例

示例1：网页截图

# 截图京东首页
python3 skills/screenshot-tool/scripts/web_screenshot.py \
  --url "https://www.jd.com" \
  --output jd_homepage.png \
  --wait 5

示例2：PPT 转图片

# 转换整个 PPT
python3 skills/screenshot-tool/scripts/doc_screenshot.py \
  --input presentation.pptx \
  --output-dir ./slides \
  --dpi 300

示例3：PDF 转图片

# 转换 PDF 前5页
python3 skills/screenshot-tool/scripts/doc_screenshot.py \
  --input document.pdf \
  --output-dir ./pages \
  --first-page 1 \
  --last-page 5

流程说明

文档转图片流程

PPT/Word/Excel → LibreOffice → PDF → pdf2image → PNG (300 DPI)
                                         ↑
                                    依赖: poppler-utils

网页截图流程

URL → agent-browser (headless Chrome) → Screenshot → PNG
              ↑
        依赖: agent-browser CLI

故障排除

LibreOffice 转换失败

# 检查 LibreOffice 安装
libreoffice --version

# 手动转换测试
libreoffice --headless --convert-to pdf file.pptx

pdf2image 错误

# 检查 poppler 安装
which pdftoppm pdfinfo

# 重新安装
sudo apt-get install -y poppler-utils

中文字体显示问题

# 安装中文字体
sudo apt-get install -y fonts-wqy-zenhei fonts-wqy-microhei fonts-noto-cjk

文件结构

skills/screenshot-tool/
├── SKILL.md              # 本文件
├── scripts/
│   ├── web_screenshot.py    # 网页截图脚本
│   └── doc_screenshot.py    # 文档转图片脚本
└── README.md             # 详细说明

License

MIT