Agent Browser

Automate browser navigation, element interaction, and page analysis from the command line.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/xintuchain/tongtong --skill agent-browser-xintuchain
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Agent Browser
Source: https://github.com/xintuchain/tongtong/tree/main/skills/agent-browser
Command: npx skills add https://github.com/xintuchain/tongtong --skill agent-browser-xintuchain

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables automated navigation, interaction, and analysis of web pages from the command line, removing manual clicking, typing, and repetitive UI testing tasks.

Core Features & Use Cases

  • Automated navigation and snapshotting that returns interactive element references for deterministic follow-up actions.
  • Comprehensive interaction commands to click, fill, type, select, upload, control mouse/keyboard, and manage dialogs and frames.
  • Integration-friendly outputs including JSON snapshots, screenshots, PDF export, video recording, session state save/load, and network routing for request interception.
  • Use Case: Automate logging in, filling and submitting forms, scraping structured data, reproducing UI bugs, or running end-to-end UI tests across authenticated pages.

Quick Start

Open a URL, snapshot interactive elements, then use refs to fill fields and click the submit button to verify the result.

Frequently Asked Questions about Agent Browser

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

FAQPage Schema
How do I automate headless web scraping and form filling from the command line?

Headless web scraping and form filling from the command line are automated by navigating to URLs, snapshotting interactive elements with stable refs, and executing click or fill commands. This process outputs JSON snapshots and manages session state for deterministic workflows.

Can I capture screenshots and export PDFs during browser automation?

Screenshots and PDFs can be captured during browser automation by running headless or headed workflows. The automation supports taking screenshots, exporting PDFs, and recording video sessions while navigating modern websites and interacting with page elements.

How does automated UI testing handle element interaction and session state?

Automated UI testing handles element interaction by snapshotting interactive elements to generate stable references for deterministic follow-up actions. It manages session state through save and load capabilities, allowing authenticated pages to maintain context across multiple testing steps.

Do I need Node.js or npm to run headless browser automation workflows?

Node.js and npm are required to run headless browser automation workflows, with optional native binaries available for extended functionality. This environment supports comprehensive interaction commands including clicking, typing, selecting, uploading, and managing dialogs or frames.

What is the best way to intercept network requests during automated web interactions?

The best way to intercept network requests during automated web interactions is using network routing capabilities for request interception. This integrates with JSON snapshot outputs and session state management to control and analyze traffic while navigating pages.

Can browser automation reproduce UI bugs across authenticated pages?

Browser automation can reproduce UI bugs across authenticated pages by saving and loading session state, navigating through login flows, and recording interactions. It captures screenshots and snapshots interactive elements to provide deterministic references for bug analysis.