agent-browser

Automate browser interactions via Chromium CDP for navigation, forms, screenshots, and data extraction.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/andreasasprou/polaris --skill agent-browser-andreasasprou
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/andreasasprou/polaris/tree/main/.agents/skills/agent-browser
Command: npx skills add https://github.com/andreasasprou/polaris --skill agent-browser-andreasasprou

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Automates repetitive and error-prone browser tasks so agents and developers can reliably navigate sites, complete forms, capture evidence, and extract page data without manual interaction.

Core Features & Use Cases

  • Interactive snapshots & refs: Discover compact element refs (@e1, @e2, ...) to perform low-token interactions like click, fill, and type.
  • Authentication & session persistence: Import browser auth, use persistent profiles or session names, and save/load state for recurring automation.
  • Capture & analysis: Take screenshots, full-page PDFs, video recordings, HARs, and profiler traces for debugging, E2E tests, and CI evidence.
  • Network control & downloads: Intercept, route, mock, or block requests, wait for downloads, and manage headers or proxies for scraping and geo-testing.
  • Security controls: Domain allowlists, action policy files, output boundaries, and configurable encryption for stored states.
  • Templates & workflows: Ready-to-use templates for form automation, authenticated sessions, and content capture.

Quick Start

Ask the skill to open the target URL, take an interactive snapshot to discover element refs, then interact with elements (fill, click, or screenshot) and save any artifacts.

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 scraping with Chromium CDP and manage authenticated sessions?

Automate web scraping and manage authenticated sessions by importing browser auth, using persistent profiles or session names, and saving or loading state for recurring automation across single-page and multi-page web applications.

Can I use interactive snapshots to fill forms and click elements during browser automation?

Interactive snapshots discover compact element refs like @e1 and @e2 to perform low-token interactions such as click, fill, and type during browser automation, enabling efficient form filling and element interaction.

What's the best way to capture full-page screenshots and video recordings for E2E testing?

The best way to capture E2E testing evidence is using configurable browser automation to take screenshots, generate full-page PDFs, record video, and capture HARs or profiler traces for debugging and CI workflows.

Does this browser automation tool support network interception and routing for web scraping?

Network interception and routing are supported for web scraping, allowing you to intercept, route, mock, or block requests, wait for downloads, and manage headers or proxies for scraping and geo-testing.

What security controls are available for browser automation and session management?

Security controls for browser automation include domain allowlists, action policy files, output boundaries, and configurable encryption for stored states to ensure safe session management and scraping operations.

Why should I use compact element refs instead of traditional selectors for web scraping?

Compact element refs like @e1 and @e2 provide concise interactions that reduce token usage during web scraping and browser automation, offering a more efficient alternative to traditional CSS or XPath selectors for element targeting.