← 返回 Skills 市场
Playwright Scraper Skill 1.2.0
作者
itsjustFred
· GitHub ↗
· v1.0.0
853
总下载
0
收藏
10
当前安装
1
版本数
在 OpenClaw 中安装
/install playwright-scraper-skill-1-2-0
功能描述
Playwright-based web scraping OpenClaw Skill with anti-bot protection. Successfully tested on complex sites like Discuss.com.hk.
安全使用建议
This skill appears to do what it claims: local Playwright scripts for normal and 'stealth' scraping. Before installing or running it: (1) review the included scripts yourself—npm install will fetch Playwright and its dependencies and npx playwright install will download browser binaries; run these in an isolated environment if you are cautious; (2) be aware the stealth script intentionally modifies browser fingerprints to evade bot detection—this is the feature, but could be legally or ethically questionable depending on target sites; (3) the docs mention proxy rotation and CAPTCHA services (2captcha) as planned — those would require third-party credentials and introduce additional risk if added later; (4) no credentials are required now, and no hidden network endpoints are present, but only run code from unknown sources if you trust the author or after manual audit.
功能分析
Type: OpenClaw Skill
Name: playwright-scraper-skill-1-2-0
Version: 1.0.0
The skill is designed for web scraping using Playwright, including anti-bot features. However, the `scripts/playwright-simple.js` and `scripts/playwright-stealth.js` files allow arbitrary file writes via environment variables (`SCREENSHOT_PATH` and `SAVE_HTML`). An attacker could craft a prompt to the OpenClaw agent to set `SCREENSHOT_PATH` to a sensitive system path (e.g., `/etc/passwd` or `~/.ssh/authorized_keys`), leading to an arbitrary file write vulnerability. While the skill's code itself does not exhibit malicious intent, this vulnerability could be exploited for denial of service or potentially remote code execution if combined with other flaws, classifying it as suspicious.
能力评估
Purpose & Capability
Name/description match the included files and usage: two Playwright scripts (simple + stealth), docs, examples, and npm dependency on Playwright. No credentials or unrelated binaries are requested, which is proportionate for a scraper.
Instruction Scope
SKILL.md and scripts instruct the agent to install dependencies (npm / npx playwright install) and run local JS scrapers. The stealth script modifies navigator properties and injects init scripts to hide automation markers (expected for anti-bot evasion). The docs mention future use of proxies and CAPTCHA solvers (2captcha) but those are not implemented in code. Instructions do not read unrelated system files or exfiltrate data to external servers.
Install Mechanism
No formal install spec in registry; documentation tells users to run npm install and npx playwright install chromium. Dependencies come from the public npm registry (package-lock references registry.npmjs.org). This is expected but means npm will fetch and install packages and browser binaries to disk—run in an environment you control.
Credentials
No required environment variables or credentials are declared. Optional env vars (HEADLESS, WAIT_TIME, SCREENSHOT_PATH, SAVE_HTML, USER_AGENT) are reasonable and directly related to scraper behavior.
Persistence & Privilege
Skill does not request always:true and does not attempt to modify other skills or system-wide configs. It runs on demand and writes only local artifacts (screenshots, saved HTML) when instructed.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install playwright-scraper-skill-1-2-0 - 安装完成后,直接呼叫该 Skill 的名称或使用
/playwright-scraper-skill-1-2-0触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
**Expanded Playwright web scraping skill with anti-bot protection and tested methods.**
- Added use case matrix for matching scraper to site anti-bot level.
- Introduced detailed instructions for three scraping levels: simple, dynamic, and anti-bot protected sites.
- Documented script options, environment variables, output formats, and best practices.
- Summarized tested anti-bot techniques with site-specific advice and troubleshooting.
- Compared performance and success rates across scraping strategies and frameworks.
- Outlined planned future improvements: proxies, cookie management, CAPTCHA solving, and batch scraping.
元数据
常见问题
Playwright Scraper Skill 1.2.0 是什么?
Playwright-based web scraping OpenClaw Skill with anti-bot protection. Successfully tested on complex sites like Discuss.com.hk. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 853 次。
如何安装 Playwright Scraper Skill 1.2.0?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install playwright-scraper-skill-1-2-0」即可一键安装,无需额外配置。
Playwright Scraper Skill 1.2.0 是免费的吗?
是的,Playwright Scraper Skill 1.2.0 完全免费(开源免费),可自由下载、安装和使用。
Playwright Scraper Skill 1.2.0 支持哪些平台?
Playwright Scraper Skill 1.2.0 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 Playwright Scraper Skill 1.2.0?
由 itsjustFred(@itsjustfred)开发并维护,当前版本 v1.0.0。
推荐 Skills