What command-line invocation does js-sandbox use?
It is launched as a stdio server by pointing Claude Desktop (or any MCP host) to the built `index.js` file.
@garc33
基本信息
暂无概览
Executes JavaScript code in an isolated environment
A Model Context Protocol server that provides a secure, isolated JavaScript execution environment. It is designed for developers and AI assistants that need to run untrusted JavaScript code safely, with configurable limits on execution time and memory.
Install dependencies with npm install, then build the server with npm run build. Configure it in Claude Desktop by adding the path to build/index.js in the claude_desktop_config.json file. Use the execute_js tool to run code, optionally setting a timeout (100–30000ms) and memory limit (1MB–100MB).
execute_js tool for running JavaScript codeIt is launched as a stdio server by pointing Claude Desktop (or any MCP host) to the built `index.js` file.
It exposes one tool: `execute_js`, which accepts the required parameter `code`, and optional parameters `timeout` and `memory`.
Since MCP servers communicate over stdio, debugging is best done using the MCP Inspector, available via `npm run inspector`, which provides a browser-based debugging URL.
Node.js and npm are required for development and building, as the server is written in JavaScript and runs via `node` on the host system.
For Claude Desktop, the configuration file is located at `~/Library/Application Support/Claude/claude_desktop_config.json` on macOS or `%APPDATA%/Claude/claude_desktop_config.json` on Windows.
Turn any JavaScript module into a sandboxed MCP (Model Context Protocol) server with automatic reflection and type inference.
MCP server that gives Claude persistent project memory — roadmap, session history, and codebase context. Type continue and Claude picks up exactly where you left off
A standalone agent runner that executes tasks using MCP (Model Context Protocol) tools via Anthropic Claude, AWS BedRock and OpenAI APIs. It enables AI agents to run autonomously in cloud environments and interact with various systems securely.
This 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
Aider-MCP-Upgraded is a production-grade multi-agent AI coding system that combines Desktop Commander (DC) investigation capabilities with Aider's implementation power. Features 70%+ token reduction, modular architecture, and intelligent workflow automation through strategic agen
评论