agent-browser

Automate browser tasks with CLI commands for navigation, form filling, and screenshots.

Updated Apr 3, 2026
One-click install
npx skills add https://github.com/YunosukeYoshino/harness --skill agent-browser-yunosukeyoshino
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/YunosukeYoshino/harness/tree/main/.agents/skills/agent-browser
Command: npx skills add https://github.com/YunosukeYoshino/harness --skill agent-browser-yunosukeyoshino

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Browser automation for AI agents to interact with websites, fill forms, click controls, and capture data across sessions.

Core Features & Use Cases

  • Browser automation CLI with commands to open pages, fill forms, click, take screenshots, scrape data, and test web apps.
  • Supports multi-session isolation, authentication patterns, and data extraction workflows for QA, testing, and automation across sites.
  • Real-world use: automate login flows, data extraction, and form submissions across multiple sites.

Quick Start

Install agent-browser and run a sample to observe browser automation in action.

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 tasks like form filling and web scraping with AI agents?

Browser automation for AI agents is handled through a CLI-based engine with commands to open pages, fill forms, click controls, take screenshots, and scrape data across multiple sessions. It supports session isolation and ref-based interactions for reliable execution.

Can I use browser automation to test web apps across multiple sites with authentication?

Yes, browser automation supports testing web apps across multiple sites by utilizing multi-session isolation and authentication patterns. You can automate login flows, data extraction workflows, and form submissions across different sites without session conflicts.

What CLI commands are available for web scraping and navigating pages?

The CLI engine provides commands like open, fill, click, wait, and snapshot to interact with websites. These commands enable navigating pages, capturing screenshots, scraping data, and submitting forms through ref-based interactions.

Does browser automation support session isolation for concurrent tasks?

Yes, session isolation is a core feature of this browser automation tool. It allows AI agents to maintain independent sessions, enabling concurrent tasks like login flows and data extraction across multiple sites without cross-session interference.

What's the best way to structure assets and scripts for automated browser workflows?

Automated browser workflows support optional references, scripts, and assets directories to organize your automation files. This structure helps manage data extraction workflows, authentication patterns, and testing scripts across multiple sessions efficiently.

Why do I need a CLI-based engine for AI agent browser automation instead of direct API calls?

A CLI-based engine provides structured commands like open, fill, click, wait, and snapshot that AI agents can reliably invoke for browser automation. This approach ensures consistent ref-based interactions, session isolation, and repeatable testing workflows across web apps.