agent-browser

Automate browser navigation, form filling, clicking, screenshots, and scraping.

1|Updated Jul 13, 2026
One-click install
npx skills add https://github.com/vh2224/forge-executor --skill agent-browser-vh2224
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/vh2224/forge-executor/tree/main/src/resources/skills/agent-browser
Command: npx skills add https://github.com/vh2224/forge-executor --skill agent-browser-vh2224

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill removes the friction of manually navigating websites, letting an AI reliably open pages, inspect interfaces, fill forms, click controls, and capture results without losing context between steps.

Core Features & Use Cases

  • Guided Browser Automation: Navigate pages, interact with forms, handle tabs, frames, dialogs, and downloads, and reuse saved session state across runs.
  • Reliable Inspection and Verification: Use snapshots, refs, screenshots, diffs, and waits to precisely target elements and confirm that actions produced the intended outcome.
  • Advanced Browser Operations: Support authentication flows, proxy routing, performance profiling, video recording, semantic locators, and scripted workflows for repeatable testing or data capture.
  • Use Case: A product or QA agent can log into a web app, submit a form, verify the resulting dashboard state, and save evidence with screenshots or recordings.

Quick Start

Ask the AI to open your target website, inspect the page with a snapshot, and then fill or click the discovered elements to complete the task.

Frequently Asked Questions about agent-browser

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

FAQPage Schema
How does browser automation handle authentication and session persistence?

Browser automation handles authentication by executing login flows and saving session state. This session persistence allows you to reuse authenticated contexts across multiple runs without repeating login steps for web testing or scraping workflows.

What is the most reliable way to target elements for web testing and scraping?

The most reliable method to target elements for web testing is using snapshot-based element refs. This approach captures precise page snapshots, allowing automated scripts to interact with forms and controls accurately without relying on brittle selectors.

Can I use browser automation for multi-step workflows across multiple pages?

Yes, you can use browser automation for repeatable multi-step workflows. It supports navigation across pages and sessions, handling tabs, frames, and dialogs while maintaining context for complex data capture and web app testing.

How do I capture screenshots and record video during automated web interaction?

You capture screenshots and record video during automated web interaction by enabling the built-in recording and profiling operations. This allows a QA agent to verify resulting dashboard states and save visual evidence of form submissions or clicks.

Does automated form filling support proxy routing and performance profiling?

Yes, automated form filling and web testing support proxy routing and performance profiling. These advanced browser operations allow you to route network traffic through proxies and measure performance during scripted navigation and scraping tasks.

Why does my browser automation lose context between page navigation steps?

Browser automation loses context between navigation steps when session persistence is not enabled. By saving session state and utilizing snapshot-based element refs with proper waits, the workflow maintains reliable context across pages and sessions.