skills.rest
jjchambers80
jjchambers80COMMUNITY
Explore all skills by this author →
GitHub

agent-browser

Automates browser interactions via CLI commands using ref-based element selection from accessibility snapshots.

Updated Feb 18, 2026
Search Tags:#screenshots#cli#browser automation#web scraping#form filling#agent-browser#headless browser
One-click install
npx skills add https://github.com/jjchambers80/judgesdirectory --skill agent-browser-jjchambers80
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/jjchambers80/judgesdirectory/tree/main/.github/skills/agent-browser
Command: npx skills add https://github.com/jjchambers80/judgesdirectory --skill agent-browser-jjchambers80

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires agent-browser.

What problem does it solve? Interacting with web pages from an AI agent or terminal workflow normally requires writing Playwright scripts or configuring MCP servers. This Skill provides a Bash-based CLI for navigating pages, filling forms, clicking elements, taking screenshots, and scraping data without writing browser automation code. ## Core Features & Use Cases - Ref-Based Element Selection: Take accessibility snapshots that assign refs (@e1, @e2) to interactive elements, then click, fill, or hover them directly. - Full Browser Control: Navigate, scroll, wait for elements, capture screenshots and PDFs, and extract text, HTML, or attribute values. - Parallel Sessions: Run multiple independent browser sessions side by side for concurrent automation tasks. - Use Case: Automate a login flow by opening a URL, snapshotting the page to find the email and password fields, filling credentials via refs, clicking the sign-in button, and verifying the result with a follow-up snapshot. ## Quick Start Ask the agent to open a website with agent-browser, snapshot the interactive elements, and fill in the login form using the returned element refs.

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 interactions from the command line?▼

Use the agent-browser CLI to open a URL, run snapshot -i to list interactive elements with refs like @e1, then issue commands such as click @e1 or fill @e2 "text". Re-snapshot after navigation or DOM changes to get updated refs.

agent-browser vs Playwright MCP for web automation?▼

agent-browser uses Bash commands with ref-based selection and suits quick one-off automation and CLI workflows. Playwright MCP provides tool-based responses and deeper MCP integration, which fits complex automation built around MCP tools.

How do I install agent-browser and its browser dependency?▼

Install it globally with npm install -g agent-browser, then run agent-browser install to download Chromium. Verify the setup by checking that the agent-browser command is available on your PATH.

Can agent-browser run multiple browser sessions in parallel?▼

Yes, use the --session flag with a session name, such as agent-browser --session browser1 open <url>, to run independent browsers concurrently. List active sessions with agent-browser session list.

How do I select elements without refs in agent-browser?▼

Use semantic locators with the find command, for example agent-browser find role button click --name "Submit" or find label "Email" fill "user@example.com". This targets elements by role, text, label, or placeholder instead of snapshot refs.

Related Skills

Explore More Software Engineering Skills→
npc-livenpc-live
168

agent-browser

Automate browser interactions via Chrome/CDP using the agent-browser CLI.

Official
Intermediate
hellosimplerickhellosimplerick

agent-browser

Automate browser navigation, form filling, data extraction, and screenshot capture.

Community
Intermediate
drt0927drt0927
5

browse

Drive a headless Chromium browser to load pages, interact with elements, and verify states.

Community
Advanced
AbdullahMalik17AbdullahMalik17
9

browser-use

Automate browser navigation, form filling, and data extraction via Playwright MCP.

Community
Intermediate