agent-browser

Automate browser tasks like login, form submission, and screenshot capture via CDP.

2|1|Updated Mar 16, 2026
One-click install
npx skills add https://github.com/ByeongminLee/nextjs-claude-code --skill agent-browser-byeongminlee
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/ByeongminLee/nextjs-claude-code/tree/main/template/skills-archive/agent-browser
Command: npx skills add https://github.com/ByeongminLee/nextjs-claude-code --skill agent-browser-byeongminlee

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill enables AI agents to automate browser-based tasks such as opening websites, filling forms, clicking controls, taking screenshots, and extracting data, reducing manual repetitive work and enabling reproducible web interactions.

Core Features & Use Cases

  • Automated browsing: navigate pages, fill forms, click elements, and capture screenshots or PDFs for verification.
  • Session management: support for multiple isolated sessions and persistent state across runs.
  • Reliable element interaction: uses compact refs from snapshots to minimize prompt length and increase robustness.

Quick Start

Open a website, log in, and capture a screenshot for verification

Frequently Asked Questions about agent-browser

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

FAQPage Schema
How do I automate web form filling and login with an AI agent?

Automate web form filling and login by orchestrating open, fill, and click actions against web pages via a CLI. The skill manages stateful sessions and uses compact refs from snapshots to interact with elements reliably across multiple contexts.

Can I capture screenshots and extract data from web pages during automated browsing?

Yes, you can capture screenshots or PDFs and extract data during automated browsing. The CLI leverages CDP to navigate pages, interact with controls, and capture verification outputs across isolated, persistent sessions.

Does this web automation tool support multiple isolated sessions with persistent state?

Yes, this web automation tool supports multiple isolated sessions with persistent state across runs. It manages frame contexts and provides snapshot-refs to ensure reliable element interaction throughout various browsing tasks.

What is the best way to minimize prompt length when automating browser interactions?

The best way to minimize prompt length is using compact refs generated from page snapshots. This approach reduces prompt size while increasing the robustness of element interaction during web automation tasks.

Can I use proxies and custom engines for web automation tasks?

Yes, you can use optional proxies and engines for web automation tasks. The CLI supports these configurations alongside observability features, allowing you to customize browsing sessions and monitor automated interactions effectively.

Why use CDP for automated web app testing and data extraction?

Use CDP for automated web app testing and data extraction because it enables reliable browser control via a CLI. It supports stateful sessions, frame contexts, and snapshot-refs, ensuring reproducible web interactions and accurate element targeting.