MicroPyramid

We just shipped MCP (Model Context Protocol) support for OnlineCompiler.io. What this means: AI coding assistants like Claude, Cursor, and VS Code Copilot can now execute code in 12 programming languages directly, no SDK to install, no local setup. Just point your MCP client at a URL. Why it matters: → AI agents generate code all day. Now they can run it too in sandboxed containers with strict resource limits. → One URL works everywhere. Claude Code, Claude Desktop, Cursor, VS Code, all supported out of the box. → Setup takes 10 seconds. Literally one command: claude mcp add onlinecompiler https://lnkd.in/gdEQHTjg We built this on FastMCP with Streamable HTTP transport, running inside our existing Django ASGI process. No extra infrastructure, no session state, it just works behind our existing API. If you're building AI-powered dev tools or coding platforms, MCP is the way to give your agents real code execution capabilities. Try it: https://lnkd.in/gjMapHQ5 #MCP #AI #DevTools #CodeExecution #OnlineCompiler #claude #cursor #vscode #codex

We just launched a synchronous code execution API at OnlineCompiler.io Our API already supported async execution (with callbacks) and real-time WebSocket connections. Today we're adding the simplest option: a synchronous endpoint. POST /api/run-code-sync/ — send your code, get the output back in the same response. What it supports: - 12 languages: Python, C, C++, Java, C#, F#, Go, Rust, PHP, Ruby, Haskell, TypeScript - Sandboxed Docker containers (512MB RAM, no network, 30s timeout) - CPU time and memory usage stats in every response - Max 100KB code input, results in under a second for most programs Why this matters: The sync endpoint is a much simpler integration path. No need to set up a callback server or manage WebSocket connections. It's ideal for AI agents, CLI tools, educational platforms, and any use case where you just need to run code and get the result. Full docs with examples in cURL, Python, and Node.js: https://lnkd.in/gvyi3gBj #API #DevTools #CodeExecution #OnlineCompiler

We just relaunched OnlineCompiler.io, and it's completely rebuilt from scratch.                                                                                                                                                   As developers, we've all been there, quickly needing to test a code snippet, share a working example, or embed a runnable code block in documentation. That's exactly what OnlineCompiler.io is built for. What is it? A free online code compiler that lets you write, compile, and execute code in 12 programming languages right from your browser. No setup, no installs. Supported languages: Python, C, C++, Java, C#, F#, PHP, Ruby, Haskell, Go, Rust, and TypeScript (Deno) What makes it different? 1. Embeddable Widget: Add a fully working code editor to your website, blog, or documentation with a single script tag. Your readers can write and run code without leaving the page. 2. REST API: Integrate code execution into your own apps. Get an API key from the dashboard and start making requests. 3. Secure Execution: Every code submission runs in an isolated Docker container with resource limits, network isolation, and automatic cleanup. 4. Passwordless Login: Sign in with Google or a magic link. No passwords to remember. 5. Modern Dashboard: Manage your API keys, widgets, and view your complete execution history from a clean interface. Who is it for? - Technical writers adding code examples to documentation - Educators building interactive coding tutorials - Developers needing a quick sandbox to test snippets - SaaS platforms that want to offer code execution to their users Try it out: https://onlinecompiler.io I'd love to hear your feedback. What languages or features would you like to see next? #OnlineCompiler #DevTools #BuildInPublic #Python #Rust #WebDevelopment #API #Programming #OpenSource #EdTech