← 返回 Skills 市场
kirkraman

Browser Automation

作者 KirkRaman · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
108
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install kirk-browser
功能描述
Browser automation using SkillBoss API Hub's web scraping capability to render web pages and extract clean, readable content.
使用说明 (SKILL.md)

Browser Automation

This skill uses SkillBoss API Hub's web scraping capability to render web pages and extract clean, readable content.

This is a custom skill generated by SkillBoss to handle headless web browsing. It utilizes the SkillBoss API Hub (/v1/pilot) to navigate to websites and scrape data for the agent's context.

Usage

skillboss browser read \x3Curl> - Renders the page and returns its text content.

Features

  • Headless browser automation
  • Web page rendering
  • Clean content extraction
  • SkillBoss API integration

Requirements

  • SkillBoss API access
  • Valid API key

License

MIT

安全使用建议
This skill will send any URL you ask it to read (and the rendered page content) to https://api.heybossai.com/v1/pilot using the SKILLBOSS_API_KEY. The registry metadata failed to list that required API key — verify the key name and origin before supplying it. Consider these precautions before installing: - Only provide SKILLBOSS_API_KEY if you trust the api.heybossai.com operator and their privacy/security practices. - Avoid using the skill to fetch private/internal URLs (localhost, 169.254.169.254, internal hostnames) because that can leak secrets or enable SSRF. - If possible, restrict the agent to public URLs or add an allowlist on your side. - Inspect the code and response handling yourself; confirm the remote service's privacy policy and retention practices for scraped content. - If you cannot validate the vendor, run the skill in an isolated environment or refuse to set the API key.
功能分析
Type: OpenClaw Skill Name: kirk-browser Version: 1.0.0 The skill is a straightforward wrapper for a web scraping service provided by SkillBoss. The code in index.js takes a URL as input and sends it to a legitimate-appearing API endpoint (api.heybossai.com) to retrieve rendered markdown content. There are no signs of data exfiltration, malicious execution, or prompt injection attempts across the documentation or implementation.
能力标签
requires-sensitive-credentials
能力评估
Purpose & Capability
The skill's name and description (headless browser via SkillBoss API Hub) match the code and SKILL.md: the runtime POSTs to a /v1/pilot endpoint to render pages. However, registry metadata declares no required environment variables while both README.md, SKILL.md, and index.js require SKILLBOSS_API_KEY — a clear mismatch in declared requirements.
Instruction Scope
SKILL.md and index.js allow the agent to read arbitrary URLs and send them to api.heybossai.com for rendering. There are no in-skill restrictions on target hosts (no whitelist or internal-host protection), so using this skill could cause the agent to disclose internal endpoints or sensitive page content to the remote service (server-side request forgery / data exfiltration risk). The instructions do not limit what the agent may fetch.
Install Mechanism
No install spec and no external downloads — this is an instruction-only skill with a small JS file. That reduces supply-chain risk compared with remote installers.
Credentials
The code expects a single credential SKILLBOSS_API_KEY (documented in README and SKILL.md) but the registry metadata lists no required env vars or primary credential. Requiring an API key is reasonable for the described capability, but the metadata omission is misleading and prevents the platform/user from properly evaluating secret exposure. The skill also transmits arbitrary page content to the third party, so the API key grants that service the ability to fetch and receive possibly sensitive data.
Persistence & Privilege
Skill flags are default (not always-included). The skill does not request elevated or persistent system privileges and does not alter other skills' configs.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install kirk-browser
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /kirk-browser 触发
  4. 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
Initial release of kirk-browser skill. - Provides browser automation using SkillBoss API Hub for web scraping. - Supports rendering web pages and extracting clean, readable content. - Integrates with SkillBoss API's `/v1/pilot` endpoint. - CLI command: `skillboss browser read <url>` to fetch page text. - Requires SkillBoss API access and a valid API key.
元数据
Slug kirk-browser
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

Browser Automation 是什么?

Browser automation using SkillBoss API Hub's web scraping capability to render web pages and extract clean, readable content. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 108 次。

如何安装 Browser Automation?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install kirk-browser」即可一键安装,无需额外配置。

Browser Automation 是免费的吗?

是的,Browser Automation 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Browser Automation 支持哪些平台?

Browser Automation 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 Browser Automation?

由 KirkRaman(@kirkraman)开发并维护,当前版本 v1.0.0。

💬 留言讨论