agent-browser

Name: agent-browser
Availability: InStock
Author: npc-live

Automate browser interactions via Chrome/CDP using the agent-browser CLI.

168|13|Updated Mar 20, 2026

Search Tags:#automation #cdp #chrome #browser #screenshot #web-scraping #agent-browser

Ready to run

npx skills add https://github.com/npc-live/clawfirm --skill agent-browser-npc-live

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/npc-live/clawfirm/tree/main/app/assets/skills/agent-browser
Command: npx skills add https://github.com/npc-live/clawfirm --skill agent-browser-npc-live

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill enables automated browser interactions with external websites through the agent-browser CLI, enabling tasks like navigation, form filling, data extraction, and visual capture without manual steps.

Core Features & Use Cases

Open URLs and navigate pages using agent-browser open and related commands.
Snapshot interactive elements, interact via element refs, fill inputs, click, and wait for network idle.
Take screenshots, extract text, and run in long-running monitor-like sessions for automated testing or data collection.
Use for QA automation, data collection, and scripted web interactions across multiple sites.

Quick Start

Install agent-browser if not present, then open a URL and snapshot elements to begin automation.

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 scraping and form filling via Chrome CDP?▼

Automating web scraping and form filling via Chrome CDP involves using the agent-browser CLI to open URLs, snapshot interactive elements, and interact through element refs to fill inputs and click buttons without manual steps.

Can I take automated screenshots during web testing using Chrome DevTools Protocol?▼

Yes, automated screenshots during web testing using Chrome DevTools Protocol can be captured by running the agent-browser CLI to navigate pages and extract visual captures for QA automation or data collection.

What do I need to run browser automation tasks with CDP locally?▼

Running browser automation tasks with CDP locally requires a Chrome or Chromium environment accessible via the DevTools Protocol and the agent-browser CLI installed on your system to execute the navigation and interaction commands.

How does element interaction work when automating browser tasks via CDP?▼

Element interaction when automating browser tasks via CDP works by snapshotting interactive elements on the page, referencing them through specific element refs, and then executing targeted actions like filling inputs, clicking, and waiting for network idle.

What is the best way to automate navigation and data extraction across multiple sites?▼

The best way to automate navigation and data extraction across multiple sites is using the agent-browser CLI in long-running monitor-like sessions, which enables scripted web interactions and continuous data collection without manual intervention.

Are there limitations when using CDP for automated browser interactions?▼

Limitations when using CDP for automated browser interactions include the strict requirement for a local Chrome or Chromium environment accessible via the protocol, meaning it cannot run without the agent-browser CLI and a compatible browser setup.