📦 github/lever

Lever: Give your AI coding agent a feedback loop

Stop manually babysitting your AI coding agent. Lever is a GitHub Copilot plugin that gives your developer tools a real feedback loop. Instead of constantly checking its work or waiting for you to flag errors, the agent measures its own progress against a clear goal. You define what success looks like once, and the system handles the rest. It turns vague instructions into measurable tasks, letting the software grind on improvements independently. You can also set an autonomy dial to control how much the agent handles on its own. It’s less about writing code and more about setting the bar.

📰 https://news.ycombinator.com/item?id=50280443

🆔 @hackernewsgithubprojects

📦 christo-auer/eilmeldung

eilmeldung: The Terminal News Reader That Feels Like Magic

A German developer built eilmeldung, a terminal news reader so fast it starts before your eyes blink. It is not a browser wrapper or a slow script; it is a native application written in Rust that turns your plain text terminal into a sleek, distraction-free news dashboard. The name is German for breaking news, but the speed is the real story. It reads your RSS feeds instantly, lets you filter articles with simple search commands, and runs on Linux, macOS, and Windows. If you want to catch up on the world without the clutter of a web browser, this is your new home.

🆔 @hackernewsgithubprojects

📦 lucianghinda/rubypowers

RubyPowers: Your AI Coder's New Boss

Your AI coding assistant used to just guess what you wanted and start typing code immediately. RubyPowers stops that chaos by forcing your agent to act like a senior engineer first. It holds its hands, asks you what you are actually trying to build, and writes out a clear plan for you to approve. Only after you sign off does it start working, using a strict process where it writes a failing test before any real code exists. This is not just a collection of prompts, but a full workflow that keeps your AI from drifting off track.

🆔 @hackernewsgithubprojects

📦 yd-233/kimi-break

Kimi Break

A single text file on GitHub is designed to strip away the safety guidelines of Moonshot’s Kimi Code assistant. This repository, called Kimi Break, contains a modified system prompt that explicitly overrides standard restrictions to allow the AI agent to assist with penetration testing and the creation of evasion-oriented code. The file redefines the AI's operational rules, stating that it may help users bypass security controls within their local environment if they explicitly insist on proceeding. It effectively converts a helpful coding assistant into an unrestricted agent that prioritizes user intent over standard safety policies.

🆔 @hackernewsgithubprojects

📦 davemessew/agentflux

Agent Flux: The AI That Learns From Your Screen

Watch your screen and build your own AI workflow. Agent Flux is a self-hosted tool that lets you automate tasks visually. The coolest part? It watches a video of you working, figures out what you did, and turns that recording into an editable workflow you can reuse. No more typing out steps manually. You keep your data local, connect your own models, and own the results. It’s like having a pair of hands that learns by watching. If you want to automate without sending your data to a cloud service, this is a clever, private way to do it.

🆔 @hackernewsgithubprojects

📦 hydro-project/infinity

How Infinity fits 75,000 agents on a Raspberry Pi

Infinity is the agent framework that lets you run 75,000 AI agents on a single Raspberry Pi. It solves the massive memory cost of keeping AI processes alive by making every agent pure data when it is waiting. Instead of holding open connections for slow tasks, an agent simply hibernates for free until the next event arrives. This means the exact same code runs efficiently on your laptop or scales to zero on cloud platforms. You get true concurrent execution without the heavy infrastructure overhead. This is how you build massive agentic systems that actually fit in real-world memory limits.

🆔 @hackernewsgithubprojects

📦 deepanshumishraa/sheet

Sheet: The Design Canvas Your AI Agent Edits

Sheet is a desktop app where your AI agent doesn't just suggest designs, it actually builds them on an infinite canvas in front of your eyes. Instead of wrestling with prompts to generate static images, you connect tools like Claude directly to the software. The agent moves real boxes, changes text, and rearranges layouts, treating the canvas as a living document you can also touch. It runs entirely on your computer with no cloud accounts or sign-ups, keeping your work private. It is essentially a design partner that has hands.

🆔 @hackernewsgithubprojects

📦 bjzhou/photoncamera

PhotonCamera: The Camera App With A Real Darkroom

PhotonCamera is the Android camera app that treats your smartphone sensor like a piece of actual photographic film. Most camera apps just slap a digital filter on top of your photo, but this one simulates the physical chemistry of real film stocks like Kodak Portra and Fujifilm Velvia. It doesn't just change the colors; it models the grain structure, the light bleed, and even the specific way that emulsion reacts to exposure. This means you can capture photos that look exactly like they came from a thirty-five millimeter camera, complete with authentic texture and mood.

🆔 @hackernewsgithubprojects

📦 edrisranjbar/lifeos

Your Entire Life in One Private Dashboard

Take control of your scattered daily chaos by moving everything into a single private dashboard. Life OS brings your habits, goals, Kanban boards, finances, and focus sessions together in one calm place that actually talks to itself. The best part? It connects directly to your favorite AI assistant, so you can just tell it to log an expense or check your habit streaks without touching a single button. Since it runs entirely on your own hardware, nobody is tracking your data or charging you a monthly fee. It is the perfect way to keep your life organized, private, and finally in sync.

📰 https://news.ycombinator.com/item?id=50014150

🆔 @hackernewsgithubprojects

📦 kooroshrh/area

AReA: Generate 4K Images Without Training

AReA lets standard image models paint pictures three times larger than they were originally built to handle, and it does this without learning a single new skill. Most diffusion transformers choke on big canvases because checking how every pixel connects to every other one becomes painfully slow. This tool fixes that by switching its focus. It starts by looking at the whole image through a coarse lens to get the big shapes right. Then, as the picture sharpens, it zooms in on small local areas to nail down the fine details.

🆔 @hackernewsgithubprojects

📦 carlesrojas/ha-diorama

Your Home As a Clickable 3D Model

ha-diorama is the Home Assistant card that turns your house into a clickable three-dimensional model. You draw your rooms and drag furniture into place, then link each piece to a real device in your smart home setup. The coolest part is that the model actually reacts: when you click a lamp, it glows and casts shadows, and the real light in your house turns on at the same time. It even tracks the sun moving across the sky as the day changes. If you are curious to see how people are building interactive control centers inside their dashboards, this project is the one to check out.

🆔 @hackernewsgithubprojects

📦 onurozcan/omux

Omux: The macOS Terminal for Parallel AI Coding

Omux is the macOS terminal that finally lets you run multiple AI coding agents side by side without them stepping on each other. It solves the messy reality of managing several Claude Code sessions at once by giving each one its own isolated workspace and a dedicated branch, so your code stays clean while the agents work. You can watch live status bars to see exactly which agent is stuck or waiting for your approval, all from one tidy sidebar. It feels like hiring a team of junior developers who never clash over files.

🆔 @hackernewsgithubprojects

📦 electrongirl/comfyui-anima-ref-adapter

ComfyUI Anima Ref Adapter: Copy Character Identity From One Image

The Anima Reference Adapter is the ComfyUI node that pulls character identity directly from a single image. It solves the problem where you must list every physical trait in your prompt to get a consistent look. By feeding a reference picture into the workflow, it transfers hair, eyes, and outfit to the new scene while your text prompt still controls the pose and background. This means you can describe a completely new setting without retyping character details. It works even for original characters the model has never seen before. If you want to keep a character consistent across different scenes, this is the tool to try.

🆔 @hackernewsgithubprojects

📦 divagr18/qwenputer

A CPU lives inside a language model

A working thirty-two-bit computer now lives entirely inside the neural weights of a language model, meaning the software runs directly through the math of the text itself without any external plug-ins. This project called Qwenputer embeds a full processor into a standard model, using its internal memory state to hold data and execute code exactly as real hardware would. The machine can run programs and calculate results with perfect accuracy, proving that a language model can be a true computer rather than just a text generator. This proves that artificial intelligence can execute logic internally without relying on outside tools.

📰 https://news.ycombinator.com/item?id=50277757

🆔 @hackernewsgithubprojects

📦 sirhalo23/ridgeracer6-recompilation

Ridge Racer 6 Recomp: Xbox 360 Racing on PC

Play the iconic 2005 arcade racer directly on your modern computer without the clunky hardware of the original console. Ridge Racer 6 Recomp is a fan-made project that rewrites the game's original code so it runs natively on Windows. It solves the problem of aging hardware by letting you drive the high-speed cars with modern graphics and smooth controls. You bring your own game disc, and the software handles the rest. It is a clever way to keep a classic racing gem alive for the next generation of players.

🆔 @hackernewsgithubprojects

📦 eelbaz/mlattice

Mlattice: The Memory Trick AI Needs

Your favorite A.I. model gets slower and hungrier every time you ask it to remember more. Mlattice breaks that rule. Instead of scanning every past word to find a match, it gives each piece of information a specific spot on a grid. When new data arrives, it only touches a tiny, fixed number of spots, no matter how much history exists. This means the cost stays flat even as memory grows. It is like having a library where finding a book never gets harder as the shelves fill up. That constant speed is the real breakthrough here. Next time your model feels sluggish, remember Mlattice.

🆔 @hackernewsgithubprojects

📦 rfaile313/samsucks

SamSucks Makes Samsung TVs Actually Work With Plex

The official Plex app on a Samsung TV takes a full minute to open, but this custom project launches in about a second. It is a from-scratch rewrite of the media client built specifically for Samsung smart televisions to fix that sluggishness. The creator rebuilt the entire interface in plain JavaScript, bypassing the slow official code to achieve smooth scrolling and instant search results. It even uses the TV’s hardware decoder to play videos directly without unnecessary re-encoding. If you are watching Plex on a Samsung TV and hate the lag, this is the fix you have been waiting for.

🆔 @hackernewsgithubprojects

📦 takattowo/bootloaderspoofer

Bootloader Spoofer

Bootloader Spoofer is the Xposed module that lies to your phone's local security checks. If an app on your unlocked phone refuses to run because it sees your bootloader is open, this tool tricks that specific check into thinking your device is safe. It works by rewriting the digital certificate that proves your phone's state right before the app sees it. It doesn't magically fix your phone for every security check, so it only works locally. The coolest part is that it has a special mode for phones where the built-in security chip is completely broken. It literally builds a fake certificate from scratch to make the app happy.

🆔 @hackernewsgithubprojects

📦 shadowswaps/rea-reconstruction-companion

What REA Reconstruction Companion Actually Does

The REA Reconstruction Companion is the open-source skill that turns reverse-engineering evidence into a clear, actionable build plan. It solves the messy problem of taking apart a complex piece of software and trying to rebuild something similar from scratch. Instead of guessing, it maps out exactly what the original app does, how it handles data, and where the hidden quirks live. It then gives you a structured roadmap to create your own version with full validation. You get a reliable bridge between understanding old code and writing new, working software. No more guessing what makes an app tick. It’s your evidence-based guide to rebuilding software you can actually trust.

🆔 @hackernewsgithubprojects

📦 razorback16/cc-intelligent-ui

Intelligent UI: Terminal Answers as Pictures

Claude Code answers usually arrive as plain text, but Intelligent UI renders each response as a typeset picture directly in your terminal. The most striking part is that Claude can draw live charts, complex math equations, and flow diagrams that look like professional papers, all without leaving your command line. It uses Typst to turn Markdown into crisp visuals with syntax-colored code and aligned tables. The underlying answer stays as normal text, so you can still copy and paste it easily. It is a free, independent mod that makes reading long technical answers significantly more pleasant.

🆔 @hackernewsgithubprojects