agent-browser

Automate web and Electron browser interactions via Chrome/Chromium CDP.

88|108|Updated Apr 7, 2026
One-click install
npx skills add https://github.com/grandamenium/cortextos --skill agent-browser-grandamenium
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/grandamenium/cortextos/tree/main/templates/analyst/.claude/skills/agent-browser
Command: npx skills add https://github.com/grandamenium/cortextos --skill agent-browser-grandamenium

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

agent-browser eliminates the need for manual browser interaction by letting AI agents reliably navigate sites, perform UI actions, and extract results.

Core Features & Use Cases

  • Website navigation and UI interaction: Open pages, click buttons, fill forms, log in, and iterate through flows for QA or operational tasks.
  • Data extraction and validation: Scrape information from rendered pages and verify outcomes with screenshot-based or DOM-based evidence.
  • Desktop app automation: Automate Electron apps like VS Code, Slack, Discord, Figma, Notion, Spotify, including checking Slack unread items and messaging workflows.
  • Cloud/browser environments: Use browser automation in Vercel Sandbox or AWS Bedrock AgentCore environments when you need isolated execution.

Quick Start

Load the browser skill before running it by instructing your agent to execute: agent-browser skills get agent-browser.

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 browser tasks like form filling and login for QA testing?

Web browser automation for form filling and login is handled by driving Chrome or Chromium via CDP, letting AI agents navigate pages, click buttons, and extract data reliably. It supports QA, exploratory testing, and dogfooding workflows without manual interaction.

Can I automate Electron desktop apps like Slack, VS Code, or Discord?

Automating Electron desktop apps like Slack, VS Code, and Discord is supported directly. AI agents can interact with desktop UIs, check Slack unread items, and execute messaging workflows using the same CDP-based browser automation approach used for web pages.

How does browser automation work in cloud environments like Vercel Sandbox or AWS Bedrock AgentCore?

Browser automation in Vercel Sandbox or AWS Bedrock AgentCore uses Chrome via CDP to execute UI interactions in isolated cloud environments. This enables programmatic web navigation, data extraction, and validation tasks without local browser dependencies.

What is the best way to scrape data from rendered web pages and validate results?

Scraping data from rendered web pages and validating results is achieved through DOM-based or screenshot-based extraction. The automation agent navigates to the target page, interacts with UI elements, and captures the rendered output to verify outcomes programmatically.

Do I need to install dependencies or run a setup command before starting browser automation?

No external dependencies are required, but the skill must be pre-loaded before performing browser actions. You load it by instructing your agent to execute the command: agent-browser skills get agent-browser, which prepares the environment for Chrome CDP automation.