概要
What is 302_sandbox_mcp?
302_sandbox_mcp is an MCP (Model Context Protocol) server that provides a secure code sandbox, enabling AI assistants to safely execute arbitrary code. It offers tools for sandbox management, code execution, file operations, and command-line execution, and supports dynamic tool loading.
How to use 302_sandbox_mcp?
Install the server via the npm package @302ai/sandbox-mcp and configure it with your 302AI_API_KEY in your MCP client (e.g., Claude Desktop, Cherry Studio, or ChatWise). The server can run in stdin mode locally or be hosted as a remote HTTP server.
Key features of 302_sandbox_mcp
- Dynamic loading of tool list from remote server.
- Supports both local stdin and remote HTTP modes.
- Create, list, and destroy isolated sandboxes.
- Execute arbitrary code with one-click or run-code tool.
- Run command-line instructions inside the sandbox.
- Import and export files to and from the sandbox.
Use cases of 302_sandbox_mcp
- Safely executing code generated by an AI assistant.
- Running interactive command-line tasks in a sandboxed environment.
- Transferring files into and out of a secure sandbox for analysis.
- Automating code testing or data processing via MCP-compatible clients.
FAQ from 302_sandbox_mcp
What API key is required?
You need a 302AI_API_KEY, which you can obtain from 302.AI’s API list page.
Which MCP clients are supported?
The server works with Claude Desktop, Cherry Studio, ChatWise, and any MCP-compatible client. Example configurations for each are provided in the README.
How can I debug the server?
Use the MCP Inspector by running npm run inspector after installing dependencies.
What modes does the server support?
It supports two modes: stdin mode for local usage and a remote HTTP server mode for hosted deployments.
Is the tool list fixed or dynamic?
The tool list is dynamically loaded from a remote server, so new tools can be added without updating the client configuration.
