← 返回 Skills 市场
13671
总下载
3
收藏
59
当前安装
1
版本数
在 OpenClaw 中安装
/install openclaw-agent-browser
功能描述
Headless browser automation CLI for AI agents. Use when interacting with websites — navigating pages, filling forms, clicking buttons, taking screenshots, ex...
安全使用建议
Install only if you intentionally need agent-driven browser automation. Use an isolated browser profile, avoid attaching to your normal Chrome session, pin or verify the npm package, enable domain/action/output limits, and require explicit approval before logging in, saving auth state, using file:// access, uploading files, downloading content, or submitting account-changing forms.
功能分析
Type: OpenClaw Skill
Name: openclaw-agent-browser
Version: 1.0.0
The OpenClaw AgentSkills bundle is classified as suspicious due to the extensive and high-risk capabilities exposed by the `agent-browser` tool, as detailed in `SKILL.md` and `references/commands.md`. Specifically, the `agent-browser eval "..."` command allows the AI agent to execute arbitrary JavaScript within the browser context, enabling potential data exfiltration (e.g., cookies, local storage, sensitive DOM content) and unauthorized network requests. Furthermore, commands like `agent-browser upload`, `agent-browser download`, and `agent-browser --allow-file-access open file:///...` grant the agent broad access to the local file system, posing risks for data exfiltration or unauthorized file manipulation. While these are designed features for browser automation, they create severe vulnerabilities for prompt injection attacks against the AI agent, allowing it to perform actions far beyond its intended scope if not strictly controlled by the OpenClaw runtime.
能力评估
Purpose & Capability
Navigation, form filling, scraping, screenshots, downloads, uploads, JavaScript evaluation, and local file viewing are coherent with a browser automation CLI, but the capability set is high impact.
Instruction Scope
The trigger description is broad and includes login, form submission, downloads, testing, monitoring, and any programmatic web interaction, while sensitive actions are not clearly gated by explicit user confirmation or default policies.
Install Mechanism
The setup script globally installs an unpinned external npm package and downloads Chromium, then launches a test browser session; this is disclosed and purpose-aligned, but provenance and version pinning are not controlled in the artifact.
Credentials
Networked browser automation, screenshots, PDFs, uploads, and downloads fit the stated purpose, and optional domain allowlist, action policy, content-boundary, and output-limit environment variables are documented.
Persistence & Privilege
The command reference documents credential saving, auth replay, cookie/localStorage state files, auto-save/restore sessions, CDP attachment to existing Chrome, and local file access without strong retention, isolation, or consent guidance.
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 openclaw-agent-browser - 安装完成后，直接呼叫该 Skill 的名称或使用
/openclaw-agent-browser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
Initial release — headless browser automation CLI for AI agents
元数据
常见问题
Agent Browser 是什么？
Headless browser automation CLI for AI agents. Use when interacting with websites — navigating pages, filling forms, clicking buttons, taking screenshots, ex... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 13671 次。
如何安装 Agent Browser？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install openclaw-agent-browser」即可一键安装，无需额外配置。
Agent Browser 是免费的吗？
是的，Agent Browser 完全免费（开源免费），可自由下载、安装和使用。
Agent Browser 支持哪些平台？
Agent Browser 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 Agent Browser？
由 bodietron（@bodietron）开发并维护，当前版本 v1.0.0。
推荐 Skills