Website Screenshot Tool

Website Screenshot Automation Tool - 网站截图自动化，支持响应式截图、批量处理、定时调度、视觉对比 | Automated website screenshots with responsive capture, batch processing, scheduling, vi...

Security Scan

The skill's code, runtime instructions, and declared requirements are consistent with a website screenshot tool; it requests only Python and standard browser automation libraries and does not ask for unrelated credentials or hidden network endpoints.

Purpose & Capability

Name/description match the implementation: the package requires python3, lists playwright/selenium in requirements.txt, and implements capture, batch, responsive, compare, and scheduling features. There are no unrelated binaries or environment variables requested.

Instruction Scope

SKILL.md and README instruct installing Python deps and running the included script. The code performs network requests to arbitrary URLs provided by the user and writes screenshots and history JSON to disk. This is expected for a screenshot tool, but it means the skill can access any reachable URL (including internal hosts if the agent environment permits) and will create files in the specified output directory.

Install Mechanism

No remote arbitrary downloads in the install spec; the only install action is a Python dependency install via requirements.txt (playwright, selenium, pytest). Playwright itself requires separate browser installs (not performed automatically here), which the README documents.

Credentials

The skill does not request environment variables, credentials, or config paths. That is proportionate: browser automation and network access are sufficient for the advertised functionality.

Persistence & Privilege

The skill is not always-enabled and does not modify other skills or global agent config. It runs on demand and stores outputs in its output_dir; scheduler exposes periodic runs but this is implemented locally in the ScreenshotScheduler class (no external scheduling service or forced persistence).

Assessment

This skill appears to do what it says (browser-based screenshots) and requests only Python and browser automation packages. Before installing or running it, consider: 1) Run in an isolated environment if you are concerned about the agent reaching internal services — the tool will visit any URL you give it. 2) Playwright requires separate browser installation (README notes `playwright install chromium`); without browsers installed captures will fail or fall back to selenium. 3) The scheduler will repeatedly contact configured URLs and write files to disk—confirm you want periodic outbound requests and storage. 4) The example code contains an absolute sys.path insertion for a workspace path (/root/.openclaw/...), which is just an example and not required; still review any paths you run. If you need stronger guarantees, run the code in a sandboxed container or VM and review/modify output_dir and scheduling behavior before enabling automatic jobs.

功能

• Capture — 单页截图，支持全页面/视口截图
• Batch — 批量截图多个URL
• Responsive — 模拟多设备尺寸（Desktop/Tablet/Mobile）
• Compare — 两个网页视觉差异对比
• Schedule — 定时截图任务调度
• History — 截图历史记录与导出

使用

from scripts.website_screenshot import WebsiteScreenshot, ScreenshotScheduler

tool = WebsiteScreenshot(output_dir="screenshots")

# 单页截图
result = tool.capture("https://example.com", full_page=True)

# 响应式截图（桌面/平板/手机）
results = tool.capture_responsive("https://example.com")

# 批量截图
results = tool.capture_batch(["https://site1.com", "https://site2.com"])

# 对比两个网页
diff = tool.compare("https://site.com/v1", "https://site.com/v2")

# 定时截图
scheduler = ScreenshotScheduler(tool)
scheduler.add_job("https://example.com", interval_minutes=60)

CLI

python3 scripts/website_screenshot.py