Toolhouse MCP Server

@MCP-Mirror

Repository

Basic information

About Toolhouse MCP Server

Mirror of

#mcp-server#toolhouse#integration

Toolhouse MCP Server config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "toolhouse-community_mcp-server-toolhouse": {
      "command": "npx",
      "args": [
        "@modelcontextprotocol/inspector",
        "uv",
        "--directory",
        "/path/to/toolhouse_mcp",
        "run",
        "toolhouse-mcp"
      ]
    }
  }
}

Toolhouse MCP Server tools

Tool list not available yet

This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.

Fetch the live tool list by running this server in a temporary sandbox using the button above.

Overview

What is Toolhouse MCP Server?

Toolhouse MCP Server connects Model Context Protocol (MCP) clients—such as the Claude Desktop app—to Toolhouse’s library of tools, using Groq’s API for fast inference. It is built for developers who want to extend MCP‑compatible applications with pre‑built, customisable tools.

How to use Toolhouse MCP Server?

Clone the repository and set the required environment variables: TOOLHOUSE_API_KEY, GROQ_API_KEY, and TOOLHOUSE_BUNDLE_NAME. Then add the server to your client’s mcpServers configuration (e.g., claude_desktop_config.json) using the uv command pointing to the local folder.

Key features of Toolhouse MCP Server

Connects MCP clients to Toolhouse’s tool library
Powered by Toolhouse and Groq for fast inference
Supports custom tool bundles via the Toolhouse dashboard
Runs locally with uv (no uvx support yet)
Debuggable with the MCP Inspector

Use cases of Toolhouse MCP Server

Enhance Claude Desktop with tools for web scraping, memory, or sending email
Build AI workflows that combine multiple third‑party tools through a single MCP interface
Prototype and test custom tool bundles before production deployment

Frequently asked questions

What API keys are needed?

You need a Toolhouse API key, a Groq API key, and the name of a Toolhouse bundle.

How do I configure the server?

Add the server definition to your client’s `mcpServers` config JSON, specifying the `uv` command and the required environment variables.

Can I run it with uvx?

No. The project is not yet configured for ephemeral environments like `uvx`. Clone the repository and run locally.

What tools are available?

You define a bundle on Toolhouse (e.g., “mcp‑toolhouse”) and add tools such as “Scrape the web”, “Memory”, or “Send Email”.

How do I debug the server?

Use the MCP Inspector. Launch it with `npx @modelcontextprotocol/inspector uv --directory /path/to/toolhouse_mcp run toolhouse-mcp`.

Comments

More Other MCP servers

Hindsight

vectorize-io

Hindsight: Agent Memory That Learns

50.4K

YOUB

Connect your AI assistant to your own YOUB endurance training account. Read recorded activities with full analysis, planned sessions, race goals, wellness (HRV, sleep, resting HR) and your athletic profile. Create or mov