Claude's Open-Source Program Reaches 100k Stars

Today we're releasing the Lightcone SDK – the fastest path from idea to working computer automation, in Python or TypeScript. The most powerful SDK for computer use. Lightcone SDK – TypeScript 0.8.0 and Python 2.35.0. Full Responses API surface: 37 fields, 200k+ context support, 12 starter templates.

Someone finally built a production-grade Agent OS in pure Rust. OpenFang isn't a Python wrapper or a generic multi-agent orchestrator. Here are the specs: → 137,728 lines of code. → 14 crates. → 1,767+ passing tests. → Zero clippy warnings. Because it's Rust, the performance metrics are insane. Cold starts happen in under 200ms. It uses just 40 MB of idle memory. It includes a SQLite memory backend with vector embeddings and canonical sessions. For me, this is what battle-tested infrastructure actually looks like. GitHub Repo: https://lnkd.in/d9p5-shX

Anthropic accidentally shipped a .map file in their npm package that pointed to a zip of the full unobfuscated TypeScript source. 512,000 lines across 1,900 files. No hack, no breach, just a missing .npmignore entry. The code was mirrored, forked, and rewritten in Rust and Python within hours to avoid DMCA. It’s an absolute rabbit hole.

CI/CD has improves code quality - and GitHub Actions makes it easier than ever to get started. I just built my first CI/CD pipeline with GitHub Actions! Check out my documentation for the step-by-step process 👇 ✅ Created a Python project with pytest tests ✅ Set up GitHub Actions workflow ✅ Configured automated testing on every push ✅ Watched my first green build pass https://lnkd.in/g3b7jxaK #CICD #GitHubActions #Python #DevOps #NextWork

I got tired of waiting… Installing heavy Python packages over a slow internet connection was killing my momentum — especially with libraries like torch. So I looked for a better way and came across devpi. Set it up on our local server, and honestly — it’s been worthy. Now I spin up a new virtual environment and install gigabytes of packages in under a minute ⚡ If you’re working with Python and large ML dependencies, setting up devpi locally is 100% worth it. Sometimes the best productivity boost isn’t new tools — it’s fixing the friction. link to github: https://lnkd.in/djJMNNnr

Write portable plugins that run natively or in the browser using Python, Rust, C#, and more! 🐍🦀 https://www.fslabs.ca/sdk The Foresight Development Kit is shipping support for runtime extensions later this year. Write application code once, and automatically generate WASM and gRPC bindings. You can distribute these bindings to build your own internal or external extension ecosystem. This video shows identical extensions written in Python and Rust running in the browser!

CI/CD has improves code quality - and GitHub Actions makes it easier than ever to get started. I just built my first CI/CD pipeline with GitHub Actions! Check out my documentation for the step-by-step process 👇 ✅ Created a Python project with pytest tests ✅ Set up GitHub Actions workflow ✅ Configured automated testing on every push ✅ Watched my first green build pass Huge thanks to @NextWork for this awesome project guide. https://lnkd.in/gqvEJrP9 #CICD #GitHubActions #Python #DevOps #NextWork

Stelvio Dev Log #90 Version 0.9 shipped last week. Starting on 0.10. Next up, three things Python devs shouldn't have to fight AWS for: 🧭 Step Functions (without the ASL JSON nightmare) 🔌 WebSocket APIs (when you need real-time) 🌐 HTTP APIs (v2: cheaper, faster, saner) Let's go!

Apache Airflow 3.2 is here, bringing partitioned Dag runs and asset events, async Python support for @task and PythonOperator, and UI theming. This quick notes guide comes with code examples for every new feature to use as patterns in your own Dags. Download the guide to learn how to: ↔️ Pass timestamps between Dags scheduled based on assets without custom workarounds ⚡ Cut task runtime by running concurrent async API calls in a single @task 🎨 Flag critical production deployments by adjusting the colors in the Airflow UI using an Airflow configuration variable Link below.

The entire Claude Code source code got leaked, and it's not an April Fools' joke... wow! It happened because it was accidentally pushed into production, in a source map file as part of their public npm package. It's already been (extracted from the map file) rewritten using Python and Rust, and it's the fastest-starred repo on GitHub so far, so now you can run Claude Code CLI with Ollama or any other LLM 🤯 What's even more interesting (at least for me) is what was hidden in their source code (much more advanced than the public release), how it works behind the scenes, and what's next. Check out this deep-dive: https://lnkd.in/dFfSvbfZ