MCP Fe MCP Server

MCP Fe MCP server exists for a simple reason — assistants are far more useful when they can act on MCP Fe directly instead of describing what you should do. MCP-FE turns your browser into an active MCP node — letting agents like Claude or Cursor query live state, read user context, and trigger actions directly inside your frontend. No browser extension required.

Setting it up

Setup follows the usual MCP pattern — install or clone the server, register it in your client's configuration file, restart the client.

What the assistant can call

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

• Worker — based edge:** a browser SharedWorker (preferred) or ServiceWorker stores event history (IndexedDB) and coordinates tool calls
• Zero — Stream Policy**: No data is ever pushed automatically. Context transfer only happens when an AI agent triggers a specific tool call
• Role — Bridges the gap between the internet and the user's local browser session
• Security — Handles Bearer token authentication to ensure only authorized agents can talk to the worker
• Routing — When a tool call comes from the Agent, the Worker routes it to the correct tab or the Main Thread
• Resilience — Implements a Ping-Pong mechanism to keep the WebSocket alive even when the user isn't actively interacting with the page
• Zero-Push — It only executes logic and sends data when the Worker explicitly asks for it (the Pull Model)
• React-first — ** @mcp-fe/mcp-worker + @mcp-fe/react-tools + @mcp-fe/react-event-tracker
• Discussions — ** architecture questions and ideas belong in GitHub Discussions
• Channel — Agent type
• WebMCP — Browser's built-in agent, extensions, assistive tech
• Package — What it's for

Configuration and credentials

You will need one environment variable: YOUR_PROXY_HOST. Keep credentials in your client's env block or a secrets manager rather than in a file you might commit.

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.
• With 12 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch MCP Fe.
• MCP clients confirm each tool call by default. Leave that on until you have watched what the mcp fe mcp server does with a few real requests.

Choosing this one

This sits in the browser automation group, where several servers overlap in what they claim to do but differ sharply once you actually set them up. MCP Fe's toolset — Worker, Zero, Role and 9 more — is a fair guide to whether it matches your workflow. It is maintained by mcp-fe; worth a glance at recent repository activity before you build anything load-bearing on it.

This entry was verified against MCP Fe's own documentation before publication; SyncDev keeps the directory reviewed rather than auto-generated.