LMStudio Toolpack

Provides a local MCP toolset with Python sandbox, web search, and memory tooling for offline LLMs.

python

16

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "zihaofu245-lmstudio-toolpack": {
      "command": "python",
      "args": [
        "python-sandbox.py"
      ]
    }
  }
}

You run local MCP servers to connect your offline or private language models to a small, focused set of tools. This collection offers a Python sandbox, a web search tool, and long-term memory capabilities, all in one lightweight environment managed by a single virtual environment (venv). You can start, configure, and use these MCP servers with a simple, consistent workflow for local experimentation and development.

How to use

You interact with each MCP server through an MCP-compatible client. The servers run locally and communicate via standard input/output. Start the local servers in the foreground so your client can connect, and route your agent’s requests to the appropriate tool.

How to install

Prerequisites: ensure you have Python >= 3.13 installed and a working Python environment.

Install the MCP toolpack using the managed workflow.

uv sync

Additional sections

Notes:

This setup defaults to stdio communication. You can switch to http in global configuration if you need a remote or centralized MCP endpoint.
The Python sandbox executes code with exec() permissions. Be mindful of security when running untrusted agents.

Run the specific MCP servers locally using Python. For the Python Sandbox, start the sandbox script directly. For the Web Search tool, start the corresponding script that powers the web searching capability.