Modal Provider

The Jupyter MCP Server can execute code in a Modal cloud sandbox, providing fully isolated, on-demand containers with configurable images and secrets. Execution is routed through the code-sandboxes modal engine.

tip

For the full, engine-level credential and parameter reference, see the Modal sandbox guide in the code-sandboxes documentation.

Requirements

Install the Modal extra and authenticate with Modal:

pip install "jupyter-mcp-server[modal]"
modal token new

For local development, modal token new is usually enough because the Modal SDK loads credentials from ~/.modal.toml.

Alternatively, provide credentials through environment variables:

Environment variable	Description
`MODAL_TOKEN_ID`	Modal token id
`MODAL_TOKEN_SECRET`	Modal token secret

For environment-based auth (CI/CD, containers, hosted runners), you need both variables. Modal authenticates with a token pair (public id + secret), so providing only one value is insufficient.

If needed, export both values from your local Modal config:

python - <<'PY'
import pathlib
import tomllib

cfg = tomllib.loads(pathlib.Path("~/.modal.toml").expanduser().read_text())
profile = cfg.get("default", cfg)
token_id = profile.get("token_id")
token_secret = profile.get("token_secret")
if token_id and token_secret:
  print(f"export MODAL_TOKEN_ID={token_id}")
  print(f"export MODAL_TOKEN_SECRET={token_secret}")
else:
  raise SystemExit("Could not find token_id/token_secret in ~/.modal.toml")
PY

Configuration

Select the Modal engine with SANDBOX_VARIANT=modal:

SANDBOX_VARIANT=modal
MODAL_TOKEN_ID=ak-...
MODAL_TOKEN_SECRET=as-...

Or via the command line:

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

MCP client configuration:

{
  "mcpServers": {
    "jupyter": {
      "command": "uvx",
      "args": ["jupyter-mcp-server@latest"],
      "env": {
        "SANDBOX_VARIANT": "modal",
        "MODAL_TOKEN_ID": "ak-...",
        "MODAL_TOKEN_SECRET": "as-..."
      }
    }
  }
}

note

Each execution runs in a fresh python -c process, so kernel state does not persist across separate tool calls. Combine statements into a single cell when you need shared state.

Requirements​

Configuration​

Requirements

Configuration