How do I install the Agent Browser skill?

Run npx skillmds add zot24/agent-browser in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Agent Browser safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: makes network calls, reads secrets. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Agent Browser?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Agent Browser free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Agent Browser?

zot24 (@zot24) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Agent Browser by zot24 — Agent Skill

Name: Agent Browser
Rating: 4.2 (5 reviews)
Author: zot24

Agent Browser

Expert on browser automation with AI agents using Vercel's agent-browser CLI. Use when the user wants to automate browsers, scrape websites, capture screenshots, or build AI-powered web workflows. Triggers on mentions of agent-browser, browser automation, headless browsing, web scraping with AI.

zot24 e955fbd 11 files · 189.5 KB Updated Sep 10, 2026

File contents

Agent Browser Skill

Expert at browser automation for AI agents using Vercel's agent-browser CLI.

Overview

agent-browser is a headless browser automation CLI optimized for AI agents with:

50+ commands for navigation, forms, screenshots, network, and storage
AI-optimized snapshots with element refs for deterministic selection
Multi-session support for isolated browser instances
Cross-platform native Rust CLI with Node.js fallback

Quick Start

npm install -g agent-browser
agent-browser install

agent-browser open example.com
agent-browser snapshot -i          # Get interactive elements with refs
agent-browser click @e2            # Click using ref
agent-browser fill @e3 "text"      # Fill input using ref
agent-browser screenshot page.png
agent-browser close

Core Concepts

Refs (Recommended for AI)

Element references from snapshots provide deterministic selection:

agent-browser snapshot        # Shows: button "Submit" [ref=e2]
agent-browser click @e2       # Click using ref

Sessions

Isolated browser instances with independent state:

agent-browser --session agent1 open site.com
agent-browser --session agent2 open other.com

Documentation

For detailed information, see the reference documentation:

Installation - npm, source builds, custom browsers
Quick Start - Basic workflow and examples
Commands - Full 50+ command reference
Selectors - Refs, CSS, semantic locators
Sessions - Multi-session and authentication
Snapshots - Accessibility tree and options
Streaming - WebSocket viewport streaming
Agent Mode - AI agent integration patterns
CDP Mode - Chrome DevTools Protocol

Common Workflows

Login Flow

agent-browser open https://example.com/login
agent-browser snapshot -i
agent-browser fill @e1 "user@example.com"
agent-browser fill @e2 "password"
agent-browser click @e3
agent-browser wait navigation

Data Extraction

agent-browser open https://example.com/data
agent-browser snapshot --json > data.json
agent-browser get text ".product-title"

Screenshot Capture

agent-browser screenshot viewport.png
agent-browser screenshot full.png --full

Upstream Sources

Repository: https://github.com/vercel-labs/agent-browser
Documentation: https://agent-browser.dev/

Sync & Update

When user runs sync: fetch latest from upstream sources, update docs/ files. When user runs diff: compare current vs upstream, report changes.

---name: agent-browserdescription: Expert on browser automation with AI agents using Vercel's agent-browser CLI. Use when the user wants to automate browsers, scrape websites, capture screenshots, or build AI-powered web workflows. Triggers on mentions of agent-browser, browser automation, headless browsing, web scraping with AI.---# Agent Browser SkillExpert at browser automation for AI agents using Vercel's agent-browser CLI.## Overview**agent-browser** is a headless browser automation CLI optimized for AI agents with:- 50+ commands for navigation, forms, screenshots, network, and storage- AI-optimized snapshots with element refs for deterministic selection- Multi-session support for isolated browser instances- Cross-platform native Rust CLI with Node.js fallback## Quick Start```bashnpm install -g agent-browseragent-browser installagent-browser open example.comagent-browser snapshot -i # Get interactive elements with refsagent-browser click @e2 # Click using refagent-browser fill @e3 "text" # Fill input using refagent-browser screenshot page.pngagent-browser close```## Core Concepts### Refs (Recommended for AI)Element references from snapshots provide deterministic selection:```bashagent-browser snapshot # Shows: button "Submit" [ref=e2]agent-browser click @e2 # Click using ref```### SessionsIsolated browser instances with independent state:```bashagent-browser --session agent1 open site.comagent-browser --session agent2 open other.com```## DocumentationFor detailed information, see the reference documentation:- **[Installation](docs/installation.md)** - npm, source builds, custom browsers- **[Quick Start](docs/quick-start.md)** - Basic workflow and examples- **[Commands](docs/commands.md)** - Full 50+ command reference- **[Selectors](docs/selectors.md)** - Refs, CSS, semantic locators- **[Sessions](docs/sessions.md)** - Multi-session and authentication- **[Snapshots](docs/snapshots.md)** - Accessibility tree and options- **[Streaming](docs/streaming.md)** - WebSocket viewport streaming- **[Agent Mode](docs/agent-mode.md)** - AI agent integration patterns- **[CDP Mode](docs/cdp-mode.md)** - Chrome DevTools Protocol## Common Workflows### Login Flow```bashagent-browser open https://example.com/loginagent-browser snapshot -iagent-browser fill @e1 "user@example.com"agent-browser fill @e2 "password"agent-browser click @e3agent-browser wait navigation```### Data Extraction```bashagent-browser open https://example.com/dataagent-browser snapshot --json > data.jsonagent-browser get text ".product-title"```### Screenshot Capture```bashagent-browser screenshot viewport.pngagent-browser screenshot full.png --full```## Upstream Sources- **Repository**: https://github.com/vercel-labs/agent-browser- **Documentation**: https://agent-browser.dev/## Sync & UpdateWhen user runs `sync`: fetch latest from upstream sources, update docs/ files.When user runs `diff`: compare current vs upstream, report changes.

Frequently asked questions