Skip to main content
Glama

AgentExecMPC

by realugbun
QUICKSTART.md3.45 kB
# 🚀 AgentExecMPC Quick Start Guide This guide will get you up and running with AgentExecMPC in minutes! ## Prerequisites Before you start, make sure you have: - ✅ [Docker](https://docs.docker.com/get-docker/) installed and running - ✅ [Git](https://git-scm.com/) to clone the repository - ✅ [Node.js](https://nodejs.org/) (only needed for Claude Desktop integration) ## Installation ```bash # Clone the repository git clone <your-repo-url> cd agentexecmpc # That's it! Now choose your setup method below ``` ## Setup Options ### 🤖 Option 1: Claude Desktop Integration (Most Popular) Perfect for using AgentExecMPC directly in Claude Desktop: ```bash # 1. Start the server (builds automatically) make quick-start # 2. Configure Claude Desktop make install-claude-config # 3. Restart Claude Desktop # Look for the MCP tools icon and start using it! ``` **Test it works:** Ask Claude to "run a shell command to list files" or "execute Python code to calculate 2+2" ### 💻 Option 2: Interactive/Development Mode Perfect for testing and development: ```bash # Build the image make build # Run interactively (great for testing) make run ``` ## Daily Usage Commands ```bash # Check if your server is running make status # View server logs make logs # Stop the server make stop # Restart the server make stop && make quick-start ``` ## Troubleshooting ### Server not starting? ```bash # Clean everything and start fresh make clean make quick-start ``` ### Claude Desktop not seeing MCP tools? 1. Make sure the server is running: `make status` 2. Reinstall the config: `make install-claude-config` 3. Restart Claude Desktop completely 4. Check server logs: `make logs` ### Want to see all available commands? ```bash make help ``` ### Need to access the container directly? ```bash # Open a shell inside the running container make shell ``` ## What Can You Do? Once AgentExecMPC is running, you can: - 🐚 **Execute shell commands** safely with timeout controls - 🐍 **Run Python code** with full standard library access - 🟢 **Execute Node.js/JavaScript** with npm package support - 🔷 **Run Go code** with optimized compilation - 📦 **Install packages** via pip, npm, or go modules - 📁 **File operations** in a sandboxed workspace ## Example Prompts for Claude Desktop Try these prompts once you have Claude Desktop integration working: ``` "Run a shell command to show the current directory and list files" "Execute Python code to create a simple web scraper using requests" "Install the pandas package and create a simple data analysis script" "Write and run a Go program that calculates fibonacci numbers" "Create a Node.js script that reads a JSON file and processes the data" ``` ## Architecture Your AgentExecMPC setup includes: - 🐳 **Docker container** with Python, Node.js, and Go pre-installed - 🔒 **Sandboxed execution** environment for safety - 📁 **Workspace directory** mounted for persistent files - 🌐 **Multiple transport options** (SSE, STDIO) - ⚡ **FastMCP server** handling all the protocol details ## Need Help? - Run `make help` to see all available commands - Check `make status` and `make logs` for diagnostics - Read the full README.md for detailed documentation - All commands are designed to be self-explanatory with helpful output --- **That's it!** 🎉 You're now ready to give AI agents powerful execution capabilities through AgentExecMPC!

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/realugbun/AgentExecMCP'

If you have feedback or need assistance with the MCP directory API, please join our Discord server