How do I install the Agent Browser skill?

Run npx skillmds@latest add itgoyo/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. 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?

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

Agent Browser by itgoyo — Agent Skill

Name: Agent Browser
Author: itgoyo

Agent Browser

Browser automation CLI for AI agents using agent-browser (Vercel Labs). Fast native Rust CLI with accessibility-tree based element selection. Use when you need headless browser control from the terminal — navigate, click, fill forms, screenshot, batch commands, AI chat mode.

itgoyo f9bb538 2 files · 3.3 KB Updated Sep 17, 2026

File contents

agent-browser

Browser automation CLI for AI agents by Vercel Labs. Repo: https://github.com/vercel-labs/agent-browser

Installation

npm (recommended, already installed globally)

npm install -g agent-browser
agent-browser install   # download Chrome for Testing (first time only)

Homebrew (macOS)

brew install agent-browser
agent-browser install

Cargo (Rust)

cargo install agent-browser
agent-browser install

Update

agent-browser upgrade

Chrome is stored at: ~/.agent-browser/browsers/chrome-/

Quick Start

agent-browser open example.com
agent-browser snapshot          # accessibility tree with @eN refs
agent-browser click @e2         # click element by ref
agent-browser fill @e3 "text"   # clear + fill input by ref
agent-browser screenshot page.png
agent-browser close

Core Commands

agent-browser open <url>              # navigate (aliases: goto, navigate)
agent-browser snapshot                # accessibility tree with refs (best for AI)
agent-browser snapshot -i             # interactive snapshot (live)
agent-browser click <sel>             # click element
agent-browser fill <sel> <text>       # clear and fill
agent-browser type <sel> <text>       # type into element
agent-browser press <key>             # press key (Enter, Tab, Control+a)
agent-browser scroll <dir> [px]       # scroll up/down/left/right
agent-browser screenshot [path]       # screenshot (--full for full page)
agent-browser screenshot --annotate   # annotated with numbered labels
agent-browser eval <js>               # run JavaScript
agent-browser wait <selector>         # wait for element
agent-browser wait --text "text"      # wait for text
agent-browser wait <ms>               # wait milliseconds
agent-browser close                   # close browser

Get Info

agent-browser get text <sel>
agent-browser get html <sel>
agent-browser get value <sel>
agent-browser get title
agent-browser get url

Semantic Locators (find)

agent-browser find role button click --name "Submit"
agent-browser find text "Sign In" click
agent-browser find label "Email" fill "test@test.com"
agent-browser find nth 2 "a" text

Batch Execution

# argument mode
agent-browser batch "open https://example.com" "snapshot -i" "screenshot"

# stdin JSON mode
echo '[["open","https://example.com"],["snapshot"],["screenshot","out.png"]]' | agent-browser batch --json

AI Chat Mode

agent-browser chat "Go to example.com and take a screenshot"  # single-shot
agent-browser chat                                              # interactive REPL

Notes & Pitfalls

Always run agent-browser install once after install to download Chrome.
Chrome stored at ~/.agent-browser/browsers/
Selectors: prefer @eN refs from snapshot for AI use; CSS selectors also work.
On Linux: agent-browser install --with-deps to install system dependencies.
Requires network proxy on restricted networks (set https_proxy before install).

---name: agent-browserdescription: Browser automation CLI for AI agents using agent-browser (Vercel Labs). Fast native Rust CLI with accessibility-tree based element selection. Use when you need headless browser control from the terminal — navigate, click, fill forms, screenshot, batch commands, AI chat mode.---# agent-browserBrowser automation CLI for AI agents by Vercel Labs.Repo: https://github.com/vercel-labs/agent-browser## Installation### npm (recommended, already installed globally)```npm install -g agent-browseragent-browser install # download Chrome for Testing (first time only)```### Homebrew (macOS)```brew install agent-browseragent-browser install```### Cargo (Rust)```cargo install agent-browseragent-browser install```### Update```agent-browser upgrade```Chrome is stored at: ~/.agent-browser/browsers/chrome-<version>/## Quick Start```bashagent-browser open example.comagent-browser snapshot # accessibility tree with @eN refsagent-browser click @e2 # click element by refagent-browser fill @e3 "text" # clear + fill input by refagent-browser screenshot page.pngagent-browser close```## Core Commands```bashagent-browser open <url> # navigate (aliases: goto, navigate)agent-browser snapshot # accessibility tree with refs (best for AI)agent-browser snapshot -i # interactive snapshot (live)agent-browser click <sel> # click elementagent-browser fill <sel> <text> # clear and fillagent-browser type <sel> <text> # type into elementagent-browser press <key> # press key (Enter, Tab, Control+a)agent-browser scroll <dir> [px] # scroll up/down/left/rightagent-browser screenshot [path] # screenshot (--full for full page)agent-browser screenshot --annotate # annotated with numbered labelsagent-browser eval <js> # run JavaScriptagent-browser wait <selector> # wait for elementagent-browser wait --text "text" # wait for textagent-browser wait <ms> # wait millisecondsagent-browser close # close browser```## Get Info```bashagent-browser get text <sel>agent-browser get html <sel>agent-browser get value <sel>agent-browser get titleagent-browser get url```## Semantic Locators (find)```bashagent-browser find role button click --name "Submit"agent-browser find text "Sign In" clickagent-browser find label "Email" fill "test@test.com"agent-browser find nth 2 "a" text```## Batch Execution```bash# argument modeagent-browser batch "open https://example.com" "snapshot -i" "screenshot"# stdin JSON modeecho '[["open","https://example.com"],["snapshot"],["screenshot","out.png"]]' | agent-browser batch --json```## AI Chat Mode```bashagent-browser chat "Go to example.com and take a screenshot" # single-shotagent-browser chat # interactive REPL```## Notes & Pitfalls- Always run `agent-browser install` once after install to download Chrome.- Chrome stored at ~/.agent-browser/browsers/- Selectors: prefer @eN refs from snapshot for AI use; CSS selectors also work.- On Linux: `agent-browser install --with-deps` to install system dependencies.- Requires network proxy on restricted networks (set https_proxy before install).

Frequently asked questions