agent-browser

Automate browser navigation, form filling, clicking, and data extraction via CLI.

Updated Mar 16, 2026

Search Tags:#screenshots #forms #session-management #web-testing #ai-agents #web-scraping #browser-automation

Ready to run

npx skills add https://github.com/Zhushuaijiang/auto-company-win --skill agent-browser-zhushuaijiang

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/Zhushuaijiang/auto-company-win/tree/main/.agents/skills/agent-browser
Command: npx skills add https://github.com/Zhushuaijiang/auto-company-win --skill agent-browser-zhushuaijiang

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Browser automation tasks for AI agents to interact with websites programmatically, including navigation, form filling, clicking, taking screenshots, extracting data, testing web apps, and managing authenticated sessions.

Core Features & Use Cases

Navigate and snapshot: open URLs, capture element references, and maintain stable selectors across navigation.
Interact and automate: fill forms, click controls, take screenshots, run tests, and export data for analysis.
Use Case: automate user onboarding flows, extract page content, verify UI workflows in QA, and generate documentation from live sites.

Quick Start

Open a website, snapshot the page, and begin automating interactions with 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 browser interactions for AI agents to navigate pages and fill forms?▼

Browser automation for AI agents handles page navigation, form filling, and element clicking. Using the agent-browser CLI with a snapshot-interact workflow, you open URLs, capture element references, and execute automated website flows.

Can I maintain authenticated sessions while scraping web data or testing UI workflows?▼

Yes, web testing and scraping can maintain authenticated sessions. The automation supports named sessions and state persistence, allowing AI agents to retain login contexts across multiple navigation and interaction steps.

How do I capture screenshots and extract page content during web testing?▼

To capture screenshots and extract page content during web testing, you use the snapshot-interact workflow to open URLs and grab element references. The automation tool then takes screenshots and exports data for analysis.

What is the best way to automate user onboarding flows and generate documentation from live sites?▼

Automating user onboarding flows and generating documentation from live sites requires a snapshot-interact workflow. AI agents navigate pages, interact with controls, capture data, and export the automated sequences for reference.

Do I need to configure environment variables to use a proxy for browser automation?▼

Configuring environment variables for a proxy is optional for browser automation. The agent-browser CLI supports optional proxy configuration and environment variable setup to customize web testing and scraping environments.

How does the snapshot-interact workflow maintain stable selectors across web page navigation?▼

The snapshot-interact workflow maintains stable selectors across navigation by capturing element references during the initial page snapshot. AI agents use these references to interact with controls reliably throughout the automation sequence.