Monty Provider

The Jupyter MCP Server can execute code in Monty, a minimal, secure Python interpreter written in Rust (pydantic-monty). Monty runs a restricted subset of Python in-process with microsecond startup and no access to the host filesystem, environment, or network unless explicitly granted.

This makes it ideal for running short, LLM-generated snippets where a full container or kernel would be overkill. Execution is routed through the code-sandboxes monty engine.

tip

For the full, engine-level configuration reference, see the Monty sandbox guide in the code-sandboxes documentation.

Requirements

Install the Monty extra:

pip install "jupyter-mcp-server[monty]"

Install the sandbox extension package used by this provider:

pip install datalayer_mcp_sandboxes

No credentials or external services are required — Monty runs entirely in-process.

Configuration

Select the Monty engine with SANDBOX_VARIANT=monty:

SANDBOX_VARIANT=monty

Or via the command line:

jupyter mcp start \
  --transport streamable-http \
  --sandbox-variant monty \
  --port 4040

MCP client configuration:

{
  "mcpServers": {
    "jupyter": {
      "command": "uvx",
      "args": ["jupyter-mcp-server@latest"],
      "env": {
        "SANDBOX_VARIANT": "monty"
      }
    }
  }
}

Session state (variables, imports, definitions) persists across cell executions, just like a regular kernel.

note

Monty supports only a subset of Python. Third-party libraries (such as numpy or pandas) and rich display outputs (images, plots) are not available. For those, use the jupyter, datalayer, kaggle, colab, or modal engines.

Requirements​

Configuration​

Requirements

Configuration