功能描述

A fast Rust-based headless browser automation CLI with Node.js fallback that enables AI agents to navigate, click, type, and snapshot pages via structured co...

使用说明 (SKILL.md)

Browser Automation with agent-browser

Name: AgentBrowser
Author: kendrick-lu

Installation

npm recommended

npm install -g agent-browser
agent-browser install
agent-browser install --with-deps

From Source

git clone https://github.com/vercel-labs/agent-browser
cd agent-browser
pnpm install
pnpm build
agent-browser install

Quick start

agent-browser open \x3Curl>        # Navigate to page
agent-browser snapshot -i       # Get interactive elements with refs
agent-browser click @e1         # Click element by ref
agent-browser fill @e2 "text"   # Fill input by ref
agent-browser close             # Close browser

Core workflow

Navigate: agent-browser open \x3Curl>
Snapshot: agent-browser snapshot -i (returns elements with refs like @e1, @e2)
Interact using refs from the snapshot
Re-snapshot after navigation or significant DOM changes

Commands

Navigation

agent-browser open \x3Curl>      # Navigate to URL
agent-browser back            # Go back
agent-browser forward         # Go forward
agent-browser reload          # Reload page
agent-browser close           # Close browser

Snapshot (page analysis)

agent-browser snapshot            # Full accessibility tree
agent-browser snapshot -i         # Interactive elements only (recommended)
agent-browser snapshot -c         # Compact output
agent-browser snapshot -d 3       # Limit depth to 3
agent-browser snapshot -s "#main" # Scope to CSS selector

Interactions (use @refs from snapshot)

agent-browser click @e1           # Click
agent-browser dblclick @e1        # Double-click
agent-browser focus @e1           # Focus element
agent-browser fill @e2 "text"     # Clear and type
agent-browser type @e2 "text"     # Type without clearing
agent-browser press Enter         # Press key
agent-browser press Control+a     # Key combination
agent-browser keydown Shift       # Hold key down
agent-browser keyup Shift         # Release key
agent-browser hover @e1           # Hover
agent-browser check @e1           # Check checkbox
agent-browser uncheck @e1         # Uncheck checkbox
agent-browser select @e1 "value"  # Select dropdown
agent-browser scroll down 500     # Scroll page
agent-browser scrollintoview @e1  # Scroll element into view
agent-browser drag @e1 @e2        # Drag and drop
agent-browser upload @e1 file.pdf # Upload files

Get information

agent-browser get text @e1        # Get element text
agent-browser get html @e1        # Get innerHTML
agent-browser get value @e1       # Get input value
agent-browser get attr @e1 href   # Get attribute
agent-browser get title           # Get page title
agent-browser get url             # Get current URL
agent-browser get count ".item"   # Count matching elements
agent-browser get box @e1         # Get bounding box

Check state

agent-browser is visible @e1      # Check if visible
agent-browser is enabled @e1      # Check if enabled
agent-browser is checked @e1      # Check if checked

Screenshots & PDF

agent-browser screenshot          # Screenshot to stdout
agent-browser screenshot path.png # Save to file
agent-browser screenshot --full   # Full page
agent-browser pdf output.pdf      # Save as PDF

Video recording

agent-browser record start ./demo.webm    # Start recording (uses current URL + state)
agent-browser click @e1                   # Perform actions
agent-browser record stop                 # Stop and save video
agent-browser record restart ./take2.webm # Stop current + start new recording

Recording creates a fresh context but preserves cookies/storage from your session. If no URL is provided, it automatically returns to your current page. For smooth demos, explore first, then start recording.

Wait

agent-browser wait @e1                     # Wait for element
agent-browser wait 2000                    # Wait milliseconds
agent-browser wait --text "Success"        # Wait for text
agent-browser wait --url "/dashboard"    # Wait for URL pattern
agent-browser wait --load networkidle      # Wait for network idle
agent-browser wait --fn "window.ready"     # Wait for JS condition

Mouse control

agent-browser mouse move 100 200      # Move mouse
agent-browser mouse down left         # Press button
agent-browser mouse up left           # Release button
agent-browser mouse wheel 100         # Scroll wheel

Semantic locators (alternative to refs)

agent-browser find role button click --name "Submit"
agent-browser find text "Sign In" click
agent-browser find label "Email" fill "[email protected]"
agent-browser find first ".item" click
agent-browser find nth 2 "a" text

Browser settings

agent-browser set viewport 1920 1080      # Set viewport size
agent-browser set device "iPhone 14"      # Emulate device
agent-browser set geo 37.7749 -122.4194   # Set geolocation
agent-browser set offline on              # Toggle offline mode
agent-browser set headers '{"X-Key":"v"}' # Extra HTTP headers
agent-browser set credentials user pass   # HTTP basic auth
agent-browser set media dark              # Emulate color scheme

Cookies & Storage

agent-browser cookies                     # Get all cookies
agent-browser cookies set name value      # Set cookie
agent-browser cookies clear               # Clear cookies
agent-browser storage local               # Get all localStorage
agent-browser storage local key           # Get specific key
agent-browser storage local set k v       # Set value
agent-browser storage local clear         # Clear all

Network

agent-browser network route \x3Curl>              # Intercept requests
agent-browser network route \x3Curl> --abort      # Block requests
agent-browser network route \x3Curl> --body '{}'  # Mock response
agent-browser network unroute [url]            # Remove routes
agent-browser network requests                 # View tracked requests
agent-browser network requests --filter api    # Filter requests

Tabs & Windows

agent-browser tab                 # List tabs
agent-browser tab new [url]       # New tab
agent-browser tab 2               # Switch to tab
agent-browser tab close           # Close tab
agent-browser window new          # New window

Frames

agent-browser frame "#iframe"     # Switch to iframe
agent-browser frame main          # Back to main frame

Dialogs

agent-browser dialog accept [text]  # Accept dialog
agent-browser dialog dismiss        # Dismiss dialog

JavaScript

agent-browser eval "document.title"   # Run JavaScript

State management

agent-browser state save auth.json    # Save session state
agent-browser state load auth.json    # Load saved state

Example: Form submission

agent-browser open https://example.com/form
agent-browser snapshot -i
# Output shows: textbox "Email" [ref=e1], textbox "Password" [ref=e2], button "Submit" [ref=e3]

agent-browser fill @e1 "[email protected]"
agent-browser fill @e2 "password123"
agent-browser click @e3
agent-browser wait --load networkidle
agent-browser snapshot -i  # Check result

Example: Authentication with saved state

# Login once
agent-browser open https://app.example.com/login
agent-browser snapshot -i
agent-browser fill @e1 "username"
agent-browser fill @e2 "password"
agent-browser click @e3
agent-browser wait --url "/dashboard"
agent-browser state save auth.json

# Later sessions: load saved state
agent-browser state load auth.json
agent-browser open https://app.example.com/dashboard

Sessions (parallel browsers)

agent-browser --session test1 open site-a.com
agent-browser --session test2 open site-b.com
agent-browser session list

JSON output (for parsing)

Add --json for machine-readable output:

agent-browser snapshot -i --json
agent-browser get text @e1 --json

Debugging

agent-browser open example.com --headed              # Show browser window
agent-browser console                                # View console messages
agent-browser console --clear                        # Clear console
agent-browser errors                                 # View page errors
agent-browser errors --clear                         # Clear errors
agent-browser highlight @e1                          # Highlight element
agent-browser trace start                            # Start recording trace
agent-browser trace stop trace.zip                   # Stop and save trace
agent-browser record start ./debug.webm              # Record from current page
agent-browser record stop                            # Save recording
agent-browser --cdp 9222 snapshot                    # Connect via CDP

Troubleshooting

If the command is not found on Linux ARM64, use the full path in the bin folder.
If an element is not found, use snapshot to find the correct ref.
If the page is not loaded, add a wait command after navigation.
Use --headed to see the browser window for debugging.

Options

--session \x3Cname> uses an isolated session.
--json provides JSON output.
--full takes a full page screenshot.
--headed shows the browser window.
--timeout sets the command timeout in milliseconds.
--cdp \x3Cport> connects via Chrome DevTools Protocol.

Notes

Refs are stable per page load but change on navigation.
Always snapshot after navigation to get new refs.
Use fill instead of type for input fields to ensure existing text is cleared.

Reporting Issues

Skill issues: Open an issue at https://github.com/TheSethRose/Agent-Browser-CLI
agent-browser CLI issues: Open an issue at https://github.com/vercel-labs/agent-browser

安全使用建议

Before installing or invoking this skill: (1) Verify the package source on npm and the referenced GitHub repository (confirm the publisher and package contents match what the SKILL.md claims). (2) Prefer installing into an isolated environment (container or VM) rather than globally until provenance is confirmed. (3) Be aware the CLI can upload local files, set HTTP headers/credentials, and manipulate cookies/storage — avoid giving the agent autonomous permission to run commands that upload files or supply secrets. (4) Note metadata inconsistencies (mismatched ownerId/version and missing homepage); ask the skill author for a trustworthy source URL or signed package if you need higher assurance. (5) If you build from source, ensure you have git and pnpm installed and that their use is acceptable; these binaries were not declared in the skill metadata.

功能分析

Type: OpenClaw Skill Name: agentbrowser Version: 1.0.0 The skill bundle for 'agent-browser' provides powerful web automation capabilities that include high-risk behaviors such as arbitrary JavaScript execution (eval), session state persistence (state save/load), and access to sensitive browser data (cookies, storage) as documented in SKILL.md. While these features are aligned with the tool's stated purpose, they represent a significant attack surface for potential data exfiltration or unauthorized actions. Additionally, the skill requires broad Bash tool permissions and includes installation steps (install --with-deps) that may involve high-privilege system changes.

能力评估

ℹ Purpose & Capability

The name/description (Rust-based headless browser with Node fallback) generally fits an npm-distributed CLI wrapper. Requiring node/npm is reasonable for the documented npm installation. However there are metadata mismatches: registry ownerId/version (1.0.0) differ from _meta.json ownerId/version (0.2.0), the skill lists no homepage/source while SKILL.md references a GitHub repo (https://github.com/vercel-labs/agent-browser). The 'From Source' instructions use git and pnpm but those binaries are not declared as required. These provenance and dependency omissions reduce confidence in the package's origin and build assumptions.

ℹ Instruction Scope

SKILL.md stays within browser automation functionality (open, snapshot, click, fill, screenshot, record, etc.). It does not instruct reading arbitrary system files or hidden configs. However the CLI commands enable operations that touch local data and credentials (upload <file>, cookies/storage manipulation, set credentials, set headers), which could be used to read or transmit local files or secrets if misused. The instructions are explicit about those features but do not constrain how they're used, so an agent with autonomy could perform broad I/O using this tool.

ℹ Install Mechanism

This is an instruction-only skill (no install spec in registry). SKILL.md recommends npm install -g agent-browser (and a source build path using git/pnpm), which means installing from the public npm registry or building from GitHub. Installing globally via npm is a normal pattern for CLIs, but it performs an external network fetch and writes binaries. The instructions reference pnpm/git without declaring them as required, and the skill provides no signed homepage or verified source URL in the metadata — this is a provenance gap to verify before running the global install.

ℹ Credentials

The skill declares no required environment variables or primary credential, which matches the static metadata. At runtime the CLI supports setting credentials, headers, and HTTP auth via commands, but these are user-invoked features (not requested up front). The absence of requested env vars is reasonable, but because the CLI can accept credentials/headers and upload local files, users should be cautious about what the agent is allowed to instruct the CLI to do.

✓ Persistence & Privilege

The skill does not request always:true and is user-invocable. It is not asking for persistent platform-level privileges in the metadata. Installing the npm CLI will create a global binary, which is normal for CLI wrappers.

版本历史

v1.0.0

Agent Browser 1.0.0 - Initial release of a fast Rust-based headless browser automation CLI with Node.js fallback. - Allows agents to automate navigation, clicking, typing, form filling, and page snapshotting with simple structured commands. - Supports advanced features including screenshots, PDF and video recording, network monitoring, state management, cookies, and storage manipulation. - Includes semantic locators, mouse controls, multi-tab/window support, frames, dialogs, and direct JavaScript evaluation. - Designed for automating web interactions, data extraction, form automation, and testing web UIs.

元数据

Slug agentbrowser

版本 1.0.0

许可证 MIT-0

累计安装 5

当前安装数 4

历史版本数 1

常见问题

AgentBrowser 是什么？

A fast Rust-based headless browser automation CLI with Node.js fallback that enables AI agents to navigate, click, type, and snapshot pages via structured co... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 540 次。

如何安装 AgentBrowser？

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

AgentBrowser 是免费的吗？

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

AgentBrowser 支持哪些平台？

AgentBrowser 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 AgentBrowser？

由 Kendrick Lu（@kendrick-lu）开发并维护，当前版本 v1.0.0。

AgentBrowser