Sandbox Mcp
@pottekkat
About Sandbox Mcp
A Model Context Protocol (MCP) server that enables LLMs to run ANY code safely in isolated Docker containers.
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"sandbox-mcp": {
"command": "path/to/sandbox-mcp",
"args": [
"--stdio"
]
}
}
}Tools
No tools detected
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is Sandbox Mcp?
Sandbox Mcp is a Model Context Protocol (MCP) server that enables LLMs to run code in secure, isolated Docker containers. It gives LLMs an easy-to-use execution environment that anyone can create and configure through a simple, AI-native MCP server running locally.
How to use Sandbox Mcp?
Download the appropriate binary for your OS from the Releases page, or install via Go (go install github.com/pottekkat/sandbox-mcp/cmd/sandbox-mcp@latest). Initialize with sandbox-mcp --pull and sandbox-mcp --build (Docker must be installed and running). Then configure your MCP host (e.g., Claude Desktop or Cursor IDE) to use the binary with the --stdio argument.
Key features of Sandbox Mcp
- Runs code in secure, isolated Docker containers
- Supports multiple languages: Python, Rust, Go, JavaScript, Java, and more
- Includes a shell sandbox for Linux commands
- Provides a network-tools sandbox for diagnostics
- Allows creation of custom sandboxes (Dockerfile + JSON config)
- Pulls and builds default sandboxes automatically
Use cases of Sandbox Mcp
- Secure code execution without exposing your system
- Automatic code testing and validation in various environments
- Safely execute and check user-submitted code
- Diagnose connectivity issues with isolated network tools
- Run educational or interview coding snippets safely
FAQ from Sandbox Mcp
How do I install Sandbox Mcp?
You can download a pre-built binary from the Releases page or install via Go with go install github.com/pottekkat/sandbox-mcp/cmd/sandbox-mcp@latest.
What sandboxes are available?
Default sandboxes include shell, python, rust, network-tools, go, javascript, apisix, and java. Each provides an isolated Docker environment for its runtime.
Do I need Docker to use Sandbox Mcp?
Yes, Docker must be installed and running for Sandbox Mcp to execute code.
Can I create my own custom sandboxes?
Yes. You can add your own sandboxes in the configuration directory ($XDG_CONFIG_HOME/sandbox-mcp/sandboxes). A sandbox is a Dockerfile and a JSON configuration file.
What license does Sandbox Mcp use?
It is open-source under the MIT License.
Frequently asked questions
How do I install Sandbox Mcp?
You can download a pre-built binary from the Releases page or install via Go with `go install github.com/pottekkat/sandbox-mcp/cmd/sandbox-mcp@latest`.
What sandboxes are available?
Default sandboxes include shell, python, rust, network-tools, go, javascript, apisix, and java. Each provides an isolated Docker environment for its runtime.
Do I need Docker to use Sandbox Mcp?
Yes, Docker must be installed and running for Sandbox Mcp to execute code.
Can I create my own custom sandboxes?
Yes. You can add your own sandboxes in the configuration directory (`$XDG_CONFIG_HOME/sandbox-mcp/sandboxes`). A sandbox is a Dockerfile and a JSON configuration file.
What license does Sandbox Mcp use?
It is open-source under the MIT License.
Basic information
More Reasoning MCP servers
Deno Sandbox MCP Server
bewt85An MCP server that allows you to run TypeScript, JavaScript, and Python code in a sandbox on your local machine using the Deno® sandbox. This server provides a controlled environment for executing code with explicit permission controls.
ArduPilot MCP Server Sandbox
hfujikawa77ArduPilotドローンをAIエージェントから操作するMCPサーバーです。
🐢🚀 Node.js Sandbox MCP Server
alfonsograzianoA Node.js–based Model Context Protocol server that spins up disposable Docker containers to execute arbitrary JavaScript.
Part 1. Real-Time LangGraph Agent with MCP Tool Execution
junfanz1This project demonstrates a decoupled real-time agent architecture that connects LangGraph agents to remote tools served by custom MCP (Modular Command Protocol) servers. The architecture enables a flexible and scalable multi-agent system where each tool can be hosted independent
Unified MCP Suite
Godzilla675A suite of Model Context Protocol (MCP) servers designed to enhance AI agent capabilities. Provides tools for media search/understanding (images, video), web information retrieval, PDF generation, and PowerPoint presentation creation, enabling agents to interact with diverse data
Comments