📦 rootprint/rootprint

Rootprint: Own Your Logs

Keep your server logs on your own hardware with full-text search. Rootprint solves the problem of handing over your system data to expensive third-party services for basic tracking. It ingests your application logs and traces, then lets you dig through them with a fast, searchable interface. You get a clean dashboard to spot errors and understand exactly what broke, without any outside company reading your data. It’s a self-hosted, open-source tool that puts control back in your hands. If you want to see your own logs without renting someone else’s software, this is the tool to try.

📰 https://news.ycombinator.com/item?id=48368810

🆔 @hackernewsgithubprojects

📦 qiannianhuanxiang/dsha

Run a Full Linux Server in Your Pocket

Transform your smartphone into a full-fledged Linux server without needing root access or command line expertise. DSHA packages a complete Ubuntu operating system directly inside an Android app, allowing you to run complex AI coding agents right from your screen. The most surprising aspect is that this entire environment functions without the typical performance drag found in other phone-based solutions, delivering speed improvements comparable to dedicated hardware. It handles everything from file management to running native applications, ensuring your data remains safe even if the app is removed. This project turns a simple phone into a powerful, portable workspace.

🆔 @hackernewsgithubprojects

📦 morten-buur/nordef-matrix-open-control

Nordef Matrix Open Control

Nordef Matrix Open Control acts as a mechanical speed bump for artificial intelligence agents, ensuring they cannot drive themselves off a cliff. Built by Morten Büür and Eirik, this tool sits outside the model itself, meaning the A I cannot talk its way out of its own leash. It checks every action against a strict list, requiring a cryptographic human signature to proceed with anything risky. If the signature is missing or expired, the system simply stops. It also keeps a tamper-proof log of every decision, so nobody can quietly change the rules later. It is not a magic shield, but a clear, honest boundary.

🆔 @hackernewsgithubprojects

📦 ethanfel/comfyui-minimaxh3-context-loop

ComfyUI MiniMax H3 Context Loop

ComfyUI MiniMax H3 Context Loop is the ComfyUI pack that actually keeps motion and sound going across video joins. It builds longer stories by stitching separate generated clips together so they feel like one continuous take, not a jarring slideshow. The magic is in how it uses the ending of one clip to guide the start of the next, ensuring characters keep moving and audio flows naturally without breaking. This solves the biggest headache in AI video editing: making multi-scene projects that don't feel chopped up.

🆔 @hackernewsgithubprojects

📦 bam-bam-2/solo-skills

Solo Skills: AI Agents for Solo Founders

Solo Skills is the open-source kit that gives solo founders a full team of AI agents without hiring anyone. Built by a community leader who automated forty-nine solo tasks, this repository shares twenty-six ready-to-use agent skills that handle everything from generating product demo videos without screen recording to turning meeting transcripts into structured notes. Each skill acts as a step-by-step procedure that AI coding assistants can read and execute, turning repetitive busywork into automated workflows. If you run a one-person business, this is a practical way to reclaim your time by letting AI handle the operational grind so you can focus on growing your work.

🆔 @hackernewsgithubprojects

📦 openai/primegaps186

Primegaps 186

OpenAI has formalized the proof that prime numbers eventually get closer than one hundred eighty-six units apart in a machine-verifiable logic system. The project, called Primegaps 186, encodes a major number theory result in Lean four, a language designed so computers can check mathematical logic. It relies on a Python script that independently verifies the heavy computational parts, ensuring the proof is not just hand-wavy but rigorously checked. While the proof assumes a few deep results as starting points, the logic connecting them to the final gap bound is fully verified. It is a fascinating look at how we use code to audit high-level math.

📰 https://news.ycombinator.com/item?id=49555257

🆔 @hackernewsgithubprojects

📦 bernardogiordano/rewind

Rewind: Your Personal Spotify Wrapped

Rewind is the self-hosted listening recap that turns your Navidrome history into a shareable visual story. It reads directly from your own music server’s database to generate a slideshow of your top songs, artists, and late-night favorites. The standout feature is its ability to export every stat card as a high-quality image, so you can post your yearly review to social media without relying on a big tech company’s data. It works right out of the box with Docker, giving you full control over your private music analytics. If you run your own music server, this is the missing piece for celebrating your listening habits in style.

📰 https://news.ycombinator.com/item?id=47702944

🆔 @hackernewsgithubprojects

📦 osfv/bravedebloater

BraveDebloater: The Safe Way to Strip Bloat

Silently strip out Brave Browser's built-in extras by applying official enterprise policies. This project tackles the messy reality that most debloating scripts hack system files or break your browser. Instead, it uses the same policy system that companies use to manage browsers, ensuring every change is official, reversible, and safe. You get a dry-run preview by default, so nothing changes until you explicitly approve it. It also creates a backup before writing anything, meaning you can restore your browser to exactly how it was if you change your mind. It refuses to disable Shields or updates, so you keep your security while losing the noise.

📰 https://news.ycombinator.com/item?id=48530920

🆔 @hackernewsgithubprojects

📦 zhaoyx239/x-translator

X-Translator: Real-Time Voice Translation

Speak in one language and hear instant translation in another without typing a single word. X-Translator bridges the gap between hearing and understanding by piping live audio into a smart pipeline that recognizes speech, converts the meaning, and synthesizes a new voice. It doesn't just read out text; it keeps track of who is speaking so the translated audio maintains distinct vocal identities for each person. You get to see the original words, the translated text, and listen to the synthesized speech all at once in your browser. It’s a neat peek into how real-time multilingual conversation is actually built.

🆔 @hackernewsgithubprojects

📦 ryze-ai-adgent/open-seo-mcp-skills

Open SEO MCP Skills: Real Data, No Estimates

Hand your AI assistant the keys to your actual Google Search Console and Analytics, and watch it run a professional SEO audit without guessing a single number. Open SEO MCP Skills connects Claude directly to your real ranking data and keyword volumes, so you get precise insights on where you are losing traffic or overpaying for ads. The most fascinating part is its ability to track exactly which pages AI engines like ChatGPT are sending people to, giving you a clear look at your visibility in AI search. It is completely free and open source, meaning you own the code and the data.

🆔 @hackernewsgithubprojects

📦 sh1nzzz/on_the_fly3r

On-the-Fly3R Builds Maps Live From Drone Videos

On-the-Fly3R builds detailed three-dimensional maps from drone video streams in real time, processing new images as they arrive without needing to re-analyze the entire footage from scratch. This framework solves the massive computing problem of scanning huge, ever-growing scenes by intelligently picking only the most relevant images to process at any given moment. It uses modern artificial intelligence models to stitch these snippets into a consistent global map, ensuring accuracy even over long distances. This makes it possible to monitor vast areas with live spatial awareness rather than waiting for a slow batch process after the fact.

🆔 @hackernewsgithubprojects

📦 facebookresearch/midtraining-distillation

Midtraining Distillation: The Logic Trick

Standard knowledge distillation during mid-training makes models better at reasoning but worse at remembering facts. Meta’s midtraining distillation project introduces a method called Switch Distillation that solves this tradeoff. It only uses the teacher model to guide the student on the hardest, most uncertain words, leaving the rest to learn normally. This keeps the reasoning gains without sacrificing factual recall. If you are building language models, this is the fix for a problem you did not know you had.

🆔 @hackernewsgithubprojects

📦 lnkiai/m3e-canvas

M3E Canvas: The Browser Sketchpad for AI Coding Prompts

Drag buttons, cards, and sliders onto a canvas to sketch out a full mobile app interface right in your browser. M3E Canvas is a design tool that lets you arrange Material Design components, link screens together, and preview the user flow before you write a single line of code. The magic happens when you are done sketching: the app translates your entire design into a concise, natural-language prompt. You copy that text, paste it into an AI coding tool, and the AI builds the actual application for you. It is a clever bridge between visual design and AI development, turning your rough ideas into working code without the usual back-and-forth.

🆔 @hackernewsgithubprojects

📦 macanalog/spicexplorer-release

SpiceXplorer Release: The Open Analog Design Workspace

Bring analog circuit design to life by spinning up a full simulation environment without installing a single library on your computer. SpiceXplorer from the MacAnalog research group gives you a Docker container that packs a real SPICE engine and three open process kits into one image. You can start exploring complex amplifier topologies or LDO stability immediately, right in your browser, with zero setup friction. It is the fastest way to see actual waveforms and performance metrics without wrestling with local dependencies. If you want to tinker with analog blocks without the usual install pain, this is your shortcut to a working lab.

🆔 @hackernewsgithubprojects

📦 boschresearch/memorydrivor

MemoryDrivoR: Driving Cars on Pure Memory

MemoryDrivoR is the self-driving research tool that proves you don't need live camera feeds to drive well. Instead of watching the road in real time, this model relies entirely on a memory of previous drives to predict safe paths. It is a clever way to test if current autonomous driving benchmarks are actually measuring real-time awareness or just pattern matching. By stripping away the need to observe live traffic, the project reveals just how much static map data and past experience can do. It is a fascinating diagnostic for the field, showing that high scores can come from memory alone rather than true perception.

🆔 @hackernewsgithubprojects

📦 chrisyrniu/isaaclab-decoupled-wbc

IsaacLab-Decoupled-WBC: Train Humanoid Robots on One GPU

IsaacLab-Decoupled-WBC trains a humanoid robot to walk and balance using just a single graphics card, which is honestly surprising for robotics work that usually demands a server rack. This framework handles the entire pipeline from training a teacher policy to distilling it into a smaller student policy that runs on actual hardware. It focuses on decoupled whole-body control, meaning the robot manages its legs and waist to stay stable while keeping its arms in fixed positions, a setup that was actually used for the Unitree G1 experiments. You can even drive the simulated robot with your keyboard to see how it responds to movement commands before deploying anything.

🆔 @hackernewsgithubprojects

📦 6lyc/recharness

How Recharness Evolves Recommendation Models

Recharness is the experimental framework that lets an AI actually improve its own recommendation code. Instead of a human manually tweaking settings, this system uses a smart guessing algorithm to pick where to look, while a language model writes the actual improvements. It tests these changes in safe, isolated sandboxes to see if they work. If a new version scores better on test data, it keeps it; if not, it throws it out and tries a different angle. This creates a self-correcting loop that finds better models faster than guessing randomly. It’s basically a tireless research assistant that iterates until it finds a winner, all within a strict time budget.

🆔 @hackernewsgithubprojects

📦 kaiwu5/awesome-ai4ai

Awesome AI4AI: The Map of Self-Improving AI

Twenty-two-three papers. That is the current count in a living catalog called Awesome AI4AI, which tracks the specific research question of whether artificial intelligence can reliably improve itself. This is not just a list of cool tech demos; it is a curated map of the field, organized into benchmarks, runtime design, and model architecture. The most surprising part is the evidence audit, which breaks down thirty-five real systems to see exactly where humans and machines take over. It turns out that while AI systems can handle execution and repair, the core goal is still owned by humans.

🆔 @hackernewsgithubprojects

📦 jonasbaeumer/herdr-file-annotator

How to review AI code changes in Herdr

The Herdr File Annotator lets an AI coding agent freeze its work and open a live review pane directly inside your terminal. This plugin solves the friction of trusting automated code changes by forcing a real human checkpoint. Instead of scrolling through endless chat logs, the agent guides you to specific code sections while you read the actual file changes. You can tag lines as fixes or questions, and that feedback flows straight back to the AI so it knows exactly what to correct.

🆔 @hackernewsgithubprojects

📦 doofzoff/simurg

Simurg: Real-Time LLM Guard

Simurg stops broken AI answers before you ever read them. When a language model starts looping or drifting into nonsense mid-sentence, this tool cuts the stream and triggers a fresh retry instantly. It works in three lines of code and runs on any standard model server. No GPU, no training, just plain Python. It watches your text as it flows, catching errors while they happen so bad data never reaches the screen. If your model is misbehaving, this is the safety net you need. Keep your users happy by filtering out the glitches automatically. Try it on your next local model deployment and see the difference in quality.

📰 https://news.ycombinator.com/item?id=49425784

🆔 @hackernewsgithubprojects

📦 microsoft/tgrep

Tgrep: The Grep That Actually Thinks

Find that one function in a massive codebase without waiting for your computer to burn out. Tgrep, made by Microsoft, skips the slow habit of scanning every single file just to find a match. Instead, it builds a smart map of your code ahead of time. While standard search tools re-read everything from scratch, this tool only looks at the specific files that could possibly contain your answer. You build the index once, keep the server running, and suddenly your search queries are nearly instant. On huge projects, it runs up to fifty times faster than the usual options.

📰 https://news.ycombinator.com/item?id=49562928

🆔 @hackernewsgithubprojects