agent-browser

Drive the agent-browser CLI to navigate pages, fill forms, and extract data.

53|33|Updated Sep 15, 2022
One-click install
npx skills add https://github.com/tutur3u/platform --skill agent-browser-tutur3u
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/tutur3u/platform/tree/main/.agents/skills/agent-browser
Command: npx skills add https://github.com/tutur3u/platform --skill agent-browser-tutur3u

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates browser interactions by driving the agent-browser CLI to navigate web pages, fill forms, capture screenshots, and extract data, reducing manual repetitive work.

Core Features & Use Cases

  • Snapshot-based element refs for deterministic interactions and reduced AI prompt load.
  • Navigate, click, fill, screenshot, and extract data across dynamic sites, enabling QA, data gathering, and automation.
  • Use Case: QA automation for form-heavy sites or data extraction workflows.

Quick Start

Open a page: agent-browser open https://example.com Snapshot: agent-browser snapshot -i Interact with elements using refs: agent-browser click @e1, agent-browser fill @e2 "text"

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 via Bash commands?

Automate browser tasks by driving the agent-browser CLI to navigate, interact with pages, and extract data. Open a page with agent-browser, take a snapshot to get element refs, then execute clicks, form filling, and screenshots across dynamic sites.

Can I fill forms and extract data from dynamic sites using a CLI?

Yes, form filling and data extraction on dynamic sites are supported through the agent-browser CLI. Use snapshot-based interactive element refs to deterministically interact with page elements, reducing prompt load for reliable QA automation workflows.

Do I need a specific environment to run agent-browser for web testing?

You need a Bash environment with the agent-browser tool installed to run web testing and automation workflows. Predefined templates and reference guides are also available to support deterministic task execution.

How do I interact with web page elements deterministically during automation?

Interact with web page elements deterministically by taking a snapshot to generate element refs. You can then execute targeted actions like clicking @e1 or filling @e2 with text, ensuring reliable and repeatable browser automation interactions.

What is the best way to capture screenshots during web testing automation?

Capture screenshots during web testing automation by driving the agent-browser CLI after navigating to the target URL. This approach enables visual validation of dynamic sites and content extraction without manual repetitive work.

Are there limitations when using agent-browser for browser automation?

Browser automation with agent-browser requires a Bash environment and the tool installed. Workflows rely on snapshot-based element refs for interactions, meaning dynamic sites must be properly loaded before executing deterministic click, fill, or extraction commands.