LLM Sandbox MCP

v0.3.43

io.github.vndee/llm-sandbox

Securely run LLM-generated code in isolated containers across 7 languages and 3 container backends.

transport:stdio runtime:pypi

View source Website

Target client

run in your project directory

BACKEND

DOCKER_HOST

KUBECONFIG

NAMESPACE

COMMIT_CONTAINER

KEEP_TEMPLATE

SANDBOX_NETWORK_MODE

SANDBOX_READ_ONLY

SANDBOX_CAP_DROP

SANDBOX_SECURITY_OPT

SANDBOX_MEMORY

SANDBOX_CPUS

Held in this page only — never stored, logged, or sent anywhere but back to your screen.

claude mcp add llm-sandbox -e BACKEND=<BACKEND> -e DOCKER_HOST=<DOCKER_HOST> -e KUBECONFIG=<KUBECONFIG> -e NAMESPACE=<NAMESPACE> -e COMMIT_CONTAINER=<COMMIT_CONTAINER> -e KEEP_TEMPLATE=<KEEP_TEMPLATE> -e SANDBOX_NETWORK_MODE=<SANDBOX_NETWORK_MODE> -e SANDBOX_READ_ONLY=<SANDBOX_READ_ONLY> -e SANDBOX_CAP_DROP=<SANDBOX_CAP_DROP> -e SANDBOX_SECURITY_OPT=<SANDBOX_SECURITY_OPT> -e SANDBOX_MEMORY=<SANDBOX_MEMORY> -e SANDBOX_CPUS=<SANDBOX_CPUS> -- uvx llm-sandbox