InnerVault

LLM Sandbox

Stale Official MCP Registry, io.github.vndee

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

Open the source (opens github.com) https://github.com/vndee/llm-sandbox

Before you use it

  • Runs on your computer with your permissions. Read what it can do before connecting it to an agent.

What it is

MCP servers from io.github.vndee.

The problem it solves and who it suits haven't been reviewed by InnerVault. The description above is the source's own.

How to get it

pypi package (stdio)

llm-sandbox@0.3.43

From the server's entry in the official MCP Registry

Settings it asks for

  • BACKEND · Container backend to use. Must match the installed extra: mcp-docker, mcp-podman, or mcp-k8s.
  • DOCKER_HOST · Docker or Podman socket URL, e.g. unix:///var/run/docker.sock
  • KUBECONFIG · Path to kubeconfig file when BACKEND=kubernetes.
  • NAMESPACE · Kubernetes namespace used for sandbox pods when BACKEND=kubernetes.
  • COMMIT_CONTAINER · Commit the container after a run so installed libraries persist between sessions.
  • KEEP_TEMPLATE · Keep the base image after the session ends to avoid re-pulling it on the next run.
  • SANDBOX_NETWORK_MODE · Network mode for the sandbox container. Set to 'none' for hardened isolation. Docker and Podman backends only.
  • SANDBOX_READ_ONLY · Mount the sandbox root filesystem read-only. Recommended: true. Docker and Podman backends only.
  • SANDBOX_CAP_DROP · Comma-separated Linux capabilities to drop. Recommended: ALL. Docker and Podman backends only.
  • SANDBOX_SECURITY_OPT · Comma-separated container security options, e.g. no-new-privileges. Docker and Podman backends only.
  • SANDBOX_MEMORY · Memory limit for the sandbox container, e.g. 4g. Docker and Podman backends only.
  • SANDBOX_CPUS · Fractional CPU allocation for the sandbox container, e.g. 1.5. Docker and Podman backends only.

Related

  1. Containarium

    MCP servers

    Open-source agent runtime — SSH-native isolation, eBPF egress policy, Kubernetes + LXC backends, GPU passthrough, MCP-native CLI

    Indexed GitHub, FootprintAI 318 ★Apache-2.0Updated today

    git clone https://github.com/FootprintAI/Containarium.git
  2. LitterBox

    MCP servers

    A self-hosted sandbox for red teams to test payloads against modern detection before deployment. MCP integration lets an LLM agent drive analysis end to end.

    Stale GitHub, BlackSnufkin 1,549 ★GPL-3.0Updated 6 months ago

    git clone https://github.com/BlackSnufkin/LitterBox.git
  3. wirken

    MCP servers

    The enterprise gateway for autonomous agents. Identity management, per-channel isolation, credential vault, per-session tamper-evident audit log.

    Stale GitHub, gebruder 212 ★MITUpdated 4 days ago

    git clone https://github.com/gebruder/wirken.git
  4. OpenRath

    MCP servers

    An open-source, PyTorch-like runtime for dynamic multi-agent and multi-session workflows.

    Stale GitHub, Rath-Team 963 ★BSD-3-ClauseUpdated 49 days ago

    git clone https://github.com/Rath-Team/OpenRath.git
  5. agentenv-framework

    MCP servers

    Creating realistic RL environments requires collaboration between researchers, engineers, and domain experts across many dimensions: artifacts, environments tools, dynamism of the environment, reproducibility, and more. There is no open source framework for building these environments effectively. Until now.

    Indexed GitHub, scaleapi 441 ★Apache-2.0Updated today

    git clone https://github.com/scaleapi/agentenv-framework.git
  6. console

    MCP servers

    Open-source, self-hosted AI gateway and control plane: OpenAI-compatible LLM routing, RAG, an MCP hub, guardrails and agent tracing for multi-tenant teams.

    Indexed GitHub, Cognipeer 27 ★AGPL-3.0Updated 4 days ago

    git clone https://github.com/Cognipeer/console.git