Coderunner
A local sandbox for your AI agents
What is Coderunner?
Coderunner is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to local sandbox for your ai agents
A local sandbox for your AI agents
This server falls under the Developer Tools category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- A local sandbox for your AI agents
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx coderunnerConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Coderunner
Coderunner is a local sandboxed MCP server that gives AI agents a safe, isolated environment to execute Python code and perform web automation tasks. Running inside Apple Container technology on Apple Silicon Macs, it provides a persistent Jupyter kernel for stateful code execution, Playwright-based web scraping, and a skill system for reusable packaged tools — all exposed as MCP tools so Claude and other AI clients can run real code without touching the host system.
Prerequisites
- Mac with Apple Silicon (M1, M2, M3, or M4 chip) running macOS
- Python 3.10 or later installed
- Git to clone the repository
- An MCP-compatible client such as Claude Desktop, OpenCode, or Gemini CLI
- OPENAI_API_KEY if using OpenAI agent integrations (optional)
Clone the repository
Clone the Coderunner repository from GitHub.
git clone https://github.com/instavm/coderunner.git
cd coderunnerRun the installer
Make the install script executable and run it. This sets up the Apple Container environment and starts the Coderunner container with the MCP server.
chmod +x install.sh
./install.shVerify the MCP server is running
The MCP server starts automatically at http://coderunner.local:8222/mcp. You can verify it is up by checking the container status.
container listConfigure Claude Desktop
Add the Coderunner MCP server to your Claude Desktop configuration file. Replace /path/to/python and /path/to/coderunner with your actual paths.
Alternatively, configure OpenCode
For OpenCode users, add the remote MCP server URL to your OpenCode config file at ~/.config/opencode/opencode.json.
{
"mcp": {
"coderunner": {
"type": "remote",
"url": "http://coderunner.local:8222/mcp"
}
}
}Coderunner Examples
Client configuration
Claude Desktop configuration pointing to the locally running Coderunner MCP server via HTTP.
{
"mcpServers": {
"coderunner": {
"type": "http",
"url": "http://coderunner.local:8222/mcp"
}
}
}Prompts to try
Example prompts to use with Coderunner's MCP tools for code execution and web scraping.
- "Write and execute Python code to generate the first 100 prime numbers"
- "Scrape the titles and URLs from the top 10 results on Hacker News"
- "List all available skills installed in Coderunner"
- "Run a Python script that reads a CSV file and outputs summary statistics"
- "Use the pdf-text-replace skill to replace text in this PDF"Troubleshooting Coderunner
Container fails to start after running install.sh
Ensure you are on an Apple Silicon Mac (M1/M2/M3/M4) with the Apple Container framework available. The installer requires macOS with Apple Virtualization Framework support.
MCP server at coderunner.local:8222 is not reachable
Check that the container is running with 'container list'. If stopped, resume it with 'container start coderunner'. If corrupted, reset with 'container delete coderunner && ./install.sh'.
Playwright web scraping tool returns errors
Chromium and Playwright are pre-installed inside the container. If issues persist, reset the container by running 'container delete coderunner && ./install.sh' to get a clean environment.
Frequently Asked Questions about Coderunner
What is Coderunner?
Coderunner is a Model Context Protocol (MCP) server that local sandbox for your ai agents It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Coderunner?
Follow the installation instructions on the Coderunner GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Coderunner?
Coderunner works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Coderunner free to use?
Yes, Coderunner is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
Coderunner Alternatives — Similar Developer Tools Servers
Looking for alternatives to Coderunner? Here are other popular developer tools servers you can use with Claude, Cursor, and VS Code.
Ecc
★ 188.2kThe agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Javaguide
★ 155.8kJava 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Gemini CLI
★ 104.5kA secure MCP server that wraps the Google Gemini CLI, allowing clients to query Gemini models using local OAuth sessions without requiring an API key. It provides tools for model interaction and diagnostics with built-in protection against command in
Awesome MCP Servers
★ 87.3k⭐ Curated list of Model Context Protocol (MCP) servers - tools that extend Claude Desktop, Cursor, Windsurf, and other MCP clients with custom capabilities.
MCP Servers
★ 86.0kModel Context Protocol Servers
CC Switch
★ 77.5kA cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Gemini CLI & Hermes Agent. Only official website: ccswitch.io
Browse More Developer Tools MCP Servers
Explore all developer tools servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Coderunner in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Coderunner?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.