skills.rest
finas
finasCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

browser-cli

Control Chrome sessions via DevTools Protocol for browser automation and debugging.

58|4|Updated Mar 13, 2026
Search Tags:#debugging#screenshots#browser automation#mcp#lighthouse#chrome devtools protocol#dom snapshot
One-click install
npx skills add https://github.com/finas/browser-cli --skill browser-cli-finas
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: browser-cli
Source: https://github.com/finas/browser-cli/tree/main
Command: npx skills add https://github.com/finas/browser-cli --skill browser-cli-finas

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill removes the friction of manual browser debugging by letting you operate a real Chrome instance from your AI assistant using Chrome DevTools Protocol commands.

Core Features & Use Cases

  • Page automation and navigation: open URLs, manage multiple tabs, and keep a persistent session across steps.
  • Visual inspection: capture screenshots and accessibility/DOM snapshots with stable element UIDs for follow-on actions.
  • Debug and diagnose web issues: collect console logs, capture network requests, run JavaScript evaluation, and perform Lighthouse/performance analysis (including mobile emulation).

Quick Start

Start Chrome with remote debugging enabled, then ask your AI assistant to take a screenshot of a target page and return the saved image path.

Frequently Asked Questions about browser-cli

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

FAQPage Schema
How do I automate Chrome debugging and take screenshots from an AI assistant?▼

Chrome debugging automation is achieved by controlling a Chrome session through the Chrome DevTools Protocol. You can navigate pages, manage tabs, take screenshots, and capture DOM snapshots directly from your AI assistant using MCP commands.

What do I need to run browser automation with Chrome DevTools Protocol via MCP?▼

Browser automation via MCP requires a running Chrome instance with remote debugging enabled on port 9222. Once this active Chrome session is established, your AI assistant can connect and execute debugging or navigation commands.

Can I capture console logs and network requests during Chrome automation?▼

Yes, Chrome automation via DevTools Protocol supports capturing console logs and network requests. You can also run JavaScript evaluation and perform Lighthouse performance analysis, including mobile emulation, during your debugging session.

How does DOM snapshotting work for reliable element clicks in browser automation?▼

DOM snapshotting in browser automation returns page IDs and UI-tree element UIDs. These stable element UIDs allow your AI assistant to reliably target and execute follow-on actions like click, hover, and fill operations on the correct elements.

What is the best way to run Lighthouse audits and performance traces on a web page via AI?▼

Running Lighthouse audits and performance traces is handled by executing audit commands through the Chrome DevTools Protocol. This allows your AI assistant to generate performance analysis reports, including mobile emulation, directly from the active Chrome session.

Related Skills

Explore More Software Engineering Skills→
mksglumksglu
43

browsirai

Connect AI agents to a running Chrome instance via the Chrome DevTools Protocol.

Community
Advanced
mpsuessermpsuesser

web-perf

Measure Core Web Vitals and diagnose page-load bottlenecks using Chrome DevTools traces.

Community
Advanced
wenermewenerme
7

chrome-devtools

Automate Chrome DevTools debugging and browser interactions via MCP.

Community
Intermediate
andreahakuandreahaku
1

browser-agent

Automate Chrome browser interactions via the DevTools Protocol using a Bun runtime.

Community
Intermediate