About browserdebug
browserdebug offers Chrome DevTools-style page debugging through a CLI and stdio MCP server. It is for deployments that provide a DevTools-protocol-compatible browser executor; the package itself does not bundle or download a browser.
Key features
- 01Record and inspect performance traces
- 02List and inspect network requests
- 03Read console messages
- 04Capture text snapshots and screenshots
- 05Emulate network, CPU, and device conditions
- 06Drive page input using element UIDs
Use cases
- 01Investigate page performance with traces
- 02Inspect network requests and console output
- 03Capture page text or screenshots
- 04Test pages under emulated conditions
MCP support
The stdio MCP server exposes 26 browser debugging and interaction tools, including navigation, performance, network, console, snapshots, screenshots, and input. A DevTools-protocol-compatible executor must be injected to control a real Chrome browser.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
26Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| uid | string | Yes | The uid of the element from a page snapshot to click |
| dblClick | boolean | No | Whether to double click the element |
Details
From the project READMEbrowserdebug
Chrome DevTools-style debugging as a CLI and a stdio MCP server in one bin. Record performance traces and read their insights, list and inspect network requests, read console messages, take text snapshots and screenshots, emulate network/CPU/device conditions, and drive page input by element uid.
The browser sits behind an injectable executor interface
(DebugBrowser in src/types.ts). This package bundles no browser
and downloads none: with no executor configured, browser commands
fail with a clear "Browser unavailable" error. Deployments inject
a DevTools-protocol-compatible executor to drive a real Chrome.
The performance-trace state machine lives in the core: a trace
must be started before it can be stopped, and cannot be started
twice.
Install
npx -y @reneturcios/browserdebug mcp
CLI
browserdebug mcp
browserdebug pages
browserdebug navigate <url>
browserdebug snapshot
browserdebug screenshot [file]
browserdebug --version
Exit codes: 0 success, 1 browser error, 2 usage error.
MCP tools
26 tools: click, drag, fill, fill_form, handle_dialog,
hover, press_key, upload_file, close_page, list_pages,
navigate_page, new_page, select_page, wait_for, emulate,
resize_page, performance_analyze_insight,
performance_start_trace, performance_stop_trace,
get_network_request, list_network_requests, evaluate_script,
get_console_message, list_console_messages, take_screenshot,
take_snapshot.
MCP client config:
{
"mcpServers": {
"browserdebug": {
"command": "npx",
"args": ["-y", "@reneturcios/browserdebug", "mcp"]
}
}
}
Docker
docker run -i ghcr.io/cloudcorprecords/browserdebug # MCP server
docker run --rm ghcr.io/cloudcorprecords/browserdebug --version
Development
npm install
npm run build
npm test
MIT license.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.