Refinery MCP

Overview

Refinery MCP is a Model Context Protocol (MCP) server: Clean raw HTML into LLM-ready text before agents spend tokens. In practice that means any MCP-compatible AI assistant can call Refinery MCP's tools directly — the model decides when to use them in a conversation or agent run.

It runs locally: the npm package @larelabs/refinery-mcp speaks MCP over stdio on your machine. Configuration happens through environment variables — you'll need APIFY_TOKEN. The current release is v0.1.6, published under the MIT license, with source at LareLabs/refinery-mcp on GitHub.

Refinery MCP is listed under AI & Agents on mcp.site and works with any MCP client — Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed and the rest of the ecosystem — using the install snippets below. If you maintain Refinery MCP for LareLabs, claim this listing to verify ownership, earn the Verified badge, and keep the details current.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "refinery_mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@larelabs/refinery-mcp"
      ],
      "env": {
        "APIFY_TOKEN": "<apify-token>"
      }
    }
  }
}

Requires APIFY_TOKEN