A Go-based MCP server for interacting with the Lightpanda Browser using the Chrome DevTools Protocol (CDP).
gomcp
is an MCP server written in
Go.
It exposes tools to interact with Lightpanda Browser via CDP protocol.
flowchart LR;
A[CDP Client]-->|SSE or stdio|gomcp;
gomcp-->|CDP|B[Lightpanda browser];
gomcp
requires you to have already installed Lightpanda
Browser.
You need to install Go to build from source.
Once you have cloned the repository, build the binary with go build
.
By default, gocmp
starts a local instance of Lightpanda browser.
On the first run, you need to download the binary with the command:
$ gomcp download
The browser is stored in the user config directory.
$XDG_CONFIG_HOME/lightpanda-gomcp
or HOME/.config/lightpanda-gomcp
on
Linux, $HOME/Library/Application Support/lightpanda-gomcp
on Macosx.
You can remove the downloaded binary with gomcp cleanup
command.
You can connect on a remote browser with the option --cdp
.
$ gomcp -cdp ws://127.0.0.1:9222 stdio
You can configure gomcp
as a source for your Claude
Desktop.
Claude Desktop uses the stdio transport to connect to an MCP server.
Edit the claude_desktop_config.json
configuration file and add gomcp
as the mcp
server and restart Claude Desktop.
{
"mcpServers": {
"lightpanda": {
"command": "/path/to/gomcp",
"args": ["stdio"]
}
}
}
The model context protocol website gives a way to find claude_desktop_config.json file.
You can start gomcp
as a
stdio.
$ ./gomcp stdio
You can start gomcp
as a
SSE.
By default, the server listens to the HTTP connection at 127.0.0.1:8081
.
$ ./gomcp sse
2025/05/06 14:37:13 INFO server listening addr=127.0.0.1:8081
gomcp
is built thanks of open source projects, in particular:
High-quality screenshot capture optimized for Claude Vision API. Automatically tiles full pages into 1072x1072 chunks (1.15 megapixels) with configurable viewports and wait strategies for dynamic content.
Turn websites into datasets with Scrapezy
A MCP server to retrieve up-to-date jobs from company career sites.
Fetches content from deepwiki.com and converts it into LLM-readable markdown.
Use 3,000+ pre-built cloud tools to extract data from websites, e-commerce, social media, search engines, maps, and more
A MCP server that provides comprehensive website snapshot capabilities using Playwright. This server enables LLMs to capture and analyze web pages through structured accessibility snapshots, network monitoring, and console message collection.
Render website screenshots with ScreenshotOne
Fast, token-efficient web content extraction that converts websites to clean Markdown. Features Mozilla Readability, smart caching, polite crawling with robots.txt support, and concurrent fetching with minimal dependencies.
Web content fetching and conversion for efficient LLM usage
Hyperbrowser is the next-generation platform empowering AI agents and enabling effortless, scalable browser automation.