Plasmate MCP Server

MCP server for Plasmate - the browser engine for AI agents. Use with Claude Code, Claude Desktop, Cursor, Windsurf.

Local serverstdioPython

What is the Plasmate MCP MCP server?

Plasmate MCP MCP server exists for a simple reason — assistants are far more useful when they can act on Plasmate MCP directly instead of describing what you should do. MCP server for Plasmate - the browser engine for AI agents. Use with Claude Code, Claude Desktop, Cursor, Windsurf.

What you get

  • 16.6x fewer tokens — - SOM is dramatically smaller than DOM dumps
  • 50x faster — - No rendering engine overhead
  • 30MB memory — - vs Chrome's 300MB+
  • Structured output — - Clean JSON, not raw HTML

Setting it up

Installation goes through your MCP client rather than a global install: point it at plasmate-mcp on npm and it is fetched when the client starts. The copy-paste blocks for Claude Desktop, Claude Code and Cursor are further down this page.

What the assistant can call

Once Plasmate MCP is connected, these are the calls the assistant has available:

  • fetch_page — Fetch a URL and get the Semantic Object Model (SOM)
  • extract_text — Get clean readable text from any URL
  • open_page — Open a persistent browser session
  • click — Click elements by SOM ID
  • evaluate — Execute JavaScript in the page
  • close_page — Close a browser session

Before you rely on it

  • It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
  • MCP clients confirm each tool call by default. Leave that on until you have watched what the plasmate mcp mcp server does with a few real requests.

Choosing this one

Among the browser automation options, the useful question is rarely "what can it do" but "what does it cost you to run" — permissions, credentials, and how much of your context its toolset consumes. Plasmate MCP's toolset — fetch_page, extract_text, open_page and 3 more — is a fair guide to whether it matches your workflow. It is maintained by plasmate-labs; worth a glance at recent repository activity before you build anything load-bearing on it.

SyncDev reviews every entry in this directory against the project's own documentation before publishing, and revisits them as servers change.

Available tools

ToolWhat it does
fetch_pageFetch a URL and get the Semantic Object Model (SOM)
extract_textGet clean readable text from any URL
open_pageOpen a persistent browser session
clickClick elements by SOM ID
evaluateExecute JavaScript in the page
close_pageClose a browser session

How to install the Plasmate MCP MCP server

{
  "mcpServers": {
    "plasmate": {
      "command": "npx",
      "args": ["-y", "plasmate-mcp"]
    }
  }
}

Configuration as documented by the project. Restart the client after saving.

Example prompts to try

  • Use Plasmate MCP to fetch page.
  • Use Plasmate MCP to extract text.
  • Use Plasmate MCP to open page.

Frequently asked questions

It connects Plasmate MCP to MCP-compatible AI assistants such as Claude and Cursor, exposing 6 tools (fetch_page, extract_text, open_page, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Plasmate MCP directly.