Chrome DevTools MCP Manager for Openclaw

A powerful management skill for OpenClaw's built-in Chrome instance and the chrome-devtools-mcp integration to enable robust browser automation.

andeymei
v1.0.0
Mar 22, 2026
1
896
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install chrome-devtools-mcp-manager

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install chrome-devtools-mcp-manager using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Chrome DevTools MCP Manager?

The Chrome DevTools MCP Manager is an essential component of the Openclaw Skills ecosystem, designed to bridge the gap between AI agents and web browsers. It manages two critical parts of the automation stack: the OpenClaw built-in Chrome browser and the chrome-devtools-mcp server. By utilizing the Model Context Protocol (MCP), this skill allows developers to expose browser control tools directly to LLMs.

This integration provides a standardized way for AI agents to interact with the web, leveraging the Chrome DevTools Protocol (CDP) on a dedicated port. Whether you are building complex scrapers, automated testing suites, or AI-driven web researchers, this skill ensures your browser environment is stable, connected, and ready for high-level operations.

Chrome DevTools MCP Manager Use Cases

  • Orchestrating automated web research sessions using AI agents.
  • Managing the lifecycle of a headless or headed Chrome instance for MCP-based tasks.
  • Debugging browser-based AI workflows by verifying CDP endpoints and MCP server configurations.
  • Integrating web interaction capabilities into custom AI agent protocols.
  • Providing a stable environment for complex web queries that require persistent login states.

How Chrome DevTools MCP Manager Works

  1. The user initiates the browser through the OpenClaw browser tool using the openclaw profile.
  2. The browser starts with remote debugging enabled on a specific CDP port (18800 by default).
  3. The chrome-devtools-mcp server is launched as a stdio service, connecting to the running Chrome instance via the CDP URL.
  4. An MCP client (like mcporter or Claude Desktop) sends JSON-RPC commands to the MCP server.
  5. The MCP server translates these commands into Chrome DevTools Protocol actions to perform tasks like navigation, clicking, or snapshotting.
  6. Status checks verify that both the browser and the CDP interface are ready before executing automation scripts.

Chrome DevTools MCP Manager Setup

To get started with this skill within the Openclaw Skills framework, follow these steps:

  1. Start the Browser:
browser({
  action: "open",
  profile: "openclaw",
  url: "about:blank"
})
  1. Configure mcporter CLI:
mcporter server add chrome-devtools \
  --command "npx" \
  --args "chrome-devtools-mcp@latest,--browserUrl,http://127.0.0.1:18800,--no-usage-statistics"
  1. Verify Connectivity:
Invoke-WebRequest -Uri "http://localhost:18800/json/version"

Chrome DevTools MCP Manager Data Schema & Taxonomy

The skill organizes its operations around the Chrome DevTools Protocol metadata and MCP tool definitions.

Component Data Managed
Browser Profile Persistent user data, cookies, and local storage in the openclaw directory.
CDP Endpoint JSON metadata containing WebSocket debugger URLs and browser version info.
MCP Tools Schema definitions for actions like navigate_page, click, and take_snapshot.
Session State Temporary stdio streams for active JSON-RPC communication.

Chrome DevTools MCP Manager Advanced Features

  • Multi-tab management allowing agents to list, create, and switch between multiple open pages.
  • Accessibility tree snapshots for enhanced AI understanding of DOM structures.
  • Custom JavaScript evaluation within the browser context for advanced data extraction.
  • Support for autoConnect mode which automatically triggers the browser lifecycle when an MCP call is made.
  • Integration with specialized search skills like deepseek-web-query for complex automated workflows.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*