chrome-devtools

Automate browser tasks across multiple Chrome DevTools MCP servers.

Updated Dec 29, 2025
One-click install
npx skills add https://github.com/zeulewan/claude-code-skills --skill chrome-devtools-zeulewan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: chrome-devtools
Source: https://github.com/zeulewan/claude-code-skills/tree/main/chrome-devtools
Command: npx skills add https://github.com/zeulewan/claude-code-skills --skill chrome-devtools-zeulewan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Chrome DevTools MCP setup often blocks concurrent browser automation; this skill orchestrates five MCP servers to run parallel agents safely, reducing wait times and resource contention.

Core Features & Use Cases

  • Parallel task execution: run multiple browser automation tasks simultaneously across distinct MCP servers.
  • Auto-select available server: automatically picks the first healthy server to execute commands like list_pages, new_page, take_screenshot, and close_page.
  • Cleanup and isolation: manual cleanup steps ensure no stale pages remain after work.

Quick Start

Use the chrome-devtools skill to initialize the five MCP servers, then run:

  1. list_pages
  2. new_page("https://example.com")
  3. take_screenshot("s1.png")
  4. close_page(pageIdx)

Frequently Asked Questions about chrome-devtools

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

FAQPage Schema
How do I run multiple browser automation tasks in parallel across Chrome DevTools servers?

Parallel browser automation reduces wait times by distributing tasks across five MCP servers. Initialize the servers, then use auto-select to route commands like new_page, take_screenshot, and list_pages to the first available server, enabling concurrent page management without resource contention.

Can I automate screenshots and page navigation across multiple Chrome instances simultaneously?

Yes. This skill orchestrates five Chrome DevTools MCP servers to run screenshot capture, page navigation, and page creation tasks simultaneously. Auto-select picks a healthy server for each command, and manual cleanup ensures isolation between parallel workflows.

What's the best way to manage multiple Chrome DevTools MCP server instances?

Initialize five MCP servers upfront, then use the auto-select workflow to route browser commands deterministically. The skill handles server health checks and explicit cleanup steps to prevent stale pages and resource conflicts across concurrent agent workloads.

Do I need to manually select which Chrome DevTools server to use for each task?

No. The skill automatically selects the first healthy server for list_pages, new_page, take_screenshot, and close_page commands. Manual cleanup steps are your only required intervention to maintain isolation between parallel tasks.

What happens if a Chrome DevTools MCP server becomes unavailable during parallel automation?

The auto-select workflow skips unavailable servers and routes commands to the next healthy instance. If all five servers are down, the skill cannot execute tasks; deterministic server selection ensures predictable failover behavior across parallel agent runs.

How do I ensure browser pages don't accumulate when running multiple agents in parallel?

Parallel automation requires explicit close_page calls after task completion. The skill supports deterministic cleanup steps to purge stale pages from each MCP server, preventing resource leaks and ensuring clean isolation between concurrent workflows.