agent-browser

Control Chrome browser interactions via CDP with CLI commands.

3|3|Updated Nov 19, 2025
One-click install
npx skills add https://github.com/griptape-ai/griptape-nodes-library-standard --skill agent-browser-griptape-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/griptape-ai/griptape-nodes-library-standard/tree/main/.agents/skills/agent-browser
Command: npx skills add https://github.com/griptape-ai/griptape-nodes-library-standard --skill agent-browser-griptape-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Automates browser interactions for AI agents by providing a CLI to control Chrome/Chromium via CDP, enabling navigation, form filling, clicking, taking screenshots, and data extraction across websites and web apps.

Core Features & Use Cases

  • Deterministic, scriptable browser actions using commands like open, snapshot, fill, click, wait, screenshot, and pdf.
  • Session management, authentication patterns, and multi-site automation for testing, QA, scraping, and workflow validation.
  • Use cases include login automation, form submission, data extraction, visual verification, and end-to-end tasks on SPAs and traditional sites.

Quick Start

Open a website, snapshot to capture element refs, fill a login form, and submit to reach a dashboard.

Frequently Asked Questions about agent-browser

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

FAQPage Schema
How do I automate web tasks for AI agents using a CLI?

You can automate web tasks for AI agents using a CLI to control Chrome/Chromium via CDP, enabling navigation, form filling, clicking, screenshots, and data extraction across web apps. It provides commands like open, snapshot, fill, click, and wait to compose deterministic workflows.

How do I handle authentication patterns and manage sessions for browser automation?

Authentication patterns and session management are supported natively for browser automation. You can open a website, snapshot to capture element refs, fill a login form, and submit to reach a dashboard, maintaining session state across multi-site automation workflows.

What is the best way to extract data and perform visual verification on SPAs?

Data extraction and visual verification on SPAs are achieved by combining snapshot to capture element references with screenshot and pdf commands. This enables deterministic, scriptable browser actions to validate workflows and extract data from traditional sites and single-page applications.

Does this browser automation tool require any external dependencies?

No external dependencies are required to use this browser automation tool. It operates as a CLI requiring only a YAML frontmatter with name and description, supporting optional references, scripts, and assets to compose deterministic workflows.

Can I use this for end-to-end QA testing and web scraping workflows?

Yes, you can use this for end-to-end QA testing and web scraping workflows. It supports multi-site automation, login automation, form submission, and workflow validation through scriptable browser actions like fill, click, wait, and screenshot.

How do I capture and reference page elements during web interactions?

You capture and reference page elements during web interactions using the snapshot command. This performs on-demand capturing with snapshots, allowing you to identify element references needed to fill forms, click buttons, and navigate deterministic workflows.