agent-browser

Automate browser interactions via a CLI using Chrome DevTools Protocol and accessibility-tree snapshots.

4|Updated Jan 29, 2026
One-click install
npx skills add https://github.com/staticaland/skills --skill agent-browser-staticaland
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/staticaland/skills/tree/main/plugins/browser/skills/agent-browser
Command: npx skills add https://github.com/staticaland/skills --skill agent-browser-staticaland

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires agent-browser.

What problem does it solve? AI agents need a reliable way to interact with websites—navigating pages, filling forms, clicking buttons, and extracting data—without heavyweight frameworks like Playwright or Puppeteer. This Skill provides a fast native CLI that drives Chrome/Chromium over CDP with stable element references. ## Core Features & Use Cases - Browser Automation: Navigate pages, fill forms, click buttons, take screenshots, and scrape data using accessibility-tree snapshots with compact @eN element refs. - Specialized Workflows: Load dedicated skills for Electron desktop apps (VS Code, Slack, Discord), Slack workspace automation, exploratory testing/QA, Vercel Sandbox microVMs, and AWS Bedrock AgentCore cloud browsers. - Sessions & Observability: Persist sessions and authentication state, record video, and monitor activity through a dashboard on port 4848. - Use Case: Ask the agent to log into a web app, walk through a checkout flow, capture screenshots at each step, and report any broken elements during a QA pass. ## Quick Start Install the CLI with npm, then ask the agent to open a website, fill out the login form, and take a screenshot of the result.

Frequently Asked Questions about agent-browser

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate browser tasks with an AI agent?▼

Install agent-browser globally with npm, run agent-browser install, then load workflow instructions with agent-browser skills get core. The CLI drives Chrome via CDP and exposes accessibility-tree snapshots with element refs for reliable clicking, typing, and navigation.

What is the difference between agent-browser and Playwright or Puppeteer?▼

agent-browser is a native Rust CLI that talks to Chrome/Chromium directly over the Chrome DevTools Protocol, with no Playwright or Puppeteer dependency. It uses accessibility-tree snapshots with compact element refs instead of CSS selectors for interaction.

Can agent-browser automate Electron desktop apps like VS Code or Slack?▼

Yes. Load the Electron specialization with agent-browser skills get electron to automate apps such as VS Code, Slack, Discord, Figma, Notion, and Spotify. A separate slack skill covers workspace tasks like checking unreads and sending messages.

Does agent-browser work in cloud environments like Vercel Sandbox?▼

Yes. Load agent-browser skills get vercel-sandbox to run automation inside Vercel Sandbox microVMs, or skills get agentcore for AWS Bedrock AgentCore cloud browsers. These specialized skills document the environment-specific setup.

Why do agent-browser instructions say to run skills get core first?▼

The SKILL.md file is a discovery stub, not the usage guide. The CLI serves skill content matching the installed version, so running agent-browser skills get core ensures instructions never go stale between releases.