May 4, 2026·1 min read

Chrome MCP Operations Runbook — Iron Rules, Architecture & Troubleshooting

Operations skill for running chrome-devtools-mcp against your real Chrome at scale. Covers the proxy architecture, five iron rules (always-via-proxy, real-browser-only, no-env-proxy, never-kill-current-session, persistent-proxy), Chrome 146+ remote-debug popup workaround, multi-agent isolation guarantees, configuration recipes for ~/.mcp.json and ~/.claude/settings.json (with the 'no glob in permissions' gotcha), step-by-step troubleshooting flow, and four field notes from real incidents — port cleanup heuristics that backfire, protocol-layer hang detection, why 'newest = keep' is wrong, and why heavy pages need filePath-first take_snapshot to avoid 25k token overflow. Pairs with the 'Chrome MCP Background Proxy' script bundle.

Anonymous
Anonymous · Community
Agent ready

Safe staging for this asset

This asset is staged first. The copied prompt tells the agent to inspect the staged files and ask before activating scripts, MCP config, or global config.

Stage only · 29/100Policy: stage
Agent surface
Any MCP/CLI agent
Kind
Mcp Config
Install
Single
Trust
Trust: Community
Entrypoint
SKILL.md
Safe staging command
npx -y tokrepo@latest install 2080f98d-1be9-4757-8b71-85b296ff6d70 --target codex

Stages files first; activation requires review of the staged README and plan.

Operations skill for running chrome-devtools-mcp against your real Chrome at scale. Covers the proxy architecture, five iron rules (always-via-proxy, real-browser-only, no-env-proxy, never-kill-current-session, persistent-proxy), Chrome 146+ remote-debug popup workaround, multi-agent isolation guarantees, configuration recipes for ~/.mcp.json and ~/.claude/settings.json (with the 'no glob in permissions' gotcha), step-by-step troubleshooting flow, and four field notes from real incidents — port cleanup heuristics that backfire, protocol-layer hang detection, why 'newest = keep' is wrong, and why heavy pages need filePath-first take_snapshot to avoid 25k token overflow. Pairs with the 'Chrome MCP Background Proxy' script bundle.

Discussion

Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.

Asset graph

How this asset connects to the rest of the registry.

Related Assets

Chrome MCP Background Proxy — Fix Popups, Focus Stealing & Multi-Agent Conflicts

Persistent CDP proxy + entry script that lets chrome-devtools-mcp run against your real, logged-in Chrome without the Chrome 146+ consent popup spamming on every connection, without focus stealing (Target.activateTarget / Page.bringToFront are intercepted, createTarget is forced to background), and without request-ID / event collisions when multiple Claude Code windows or sub-agents share one Chrome. Includes the proxy core (cdp-proxy.mjs v3), entry script, safe cleanup, pre-flight healthcheck, and a launchd-style self-healing watchdog with Feishu alerts.

MCP ConfigsScripts
henuwangkai· ⭐ 1

Codex Chrome MCP Proxy v3

Sanitized public Chrome MCP bundle for Codex: persistent CDP proxy, real Chrome login-state control, background tabs, focus protection, multi-agent isolation, and cached chrome-devtools-mcp startup fallback.

ScriptsConfigs
henuwangkai

Multi-Browser MCP Proxies — Arc Browser & Chrome Beta Variants

Companion to 'Chrome MCP Background Proxy' for running parallel, isolated MCP fleets against Arc Browser and Chrome Beta on top of the same cdp-proxy.mjs. Arc-specific proxy auto-discovers the WebSocket path from /json/version (Arc doesn't write a DevToolsActivePort file in the standard location); Chrome Beta proxy points at Beta's own DevToolsActivePort. Lets you run mcp__chrome__*, mcp__beta__*, and mcp__arc__* side-by-side with independent client state and no cross-talk.

MCP Configs
henuwangkai

Chrome DevTools MCP — Browser Debugging for AI Agents

Give your AI coding agent full access to Chrome DevTools for browser automation, debugging, and performance analysis. Works with Claude, Cursor, Copilot, and 15+ AI tools.

MCP Configs
MCP Hub
◈Home🔍Search👤Me