📦 yujunzhixue/dsh-purge

dsh-purge: The Tool That Unlocks Every Model

dsh-purge is the local harness plugin that finally lets every model speak freely by stripping away the refusal locks that often come from the host layer, not the model itself. Many users blame the AI for saying no, but the platform wrapping it might be the real blocker. This tool maps those guardrails and applies reversible patches to your local setup so you can see exactly where the limits are coming from. It works with different models and lets you swap prompts per model to test behaviors without breaking your install. It is not an attack kit, but a research aid for understanding how your local environment shapes the conversation.

🆔 @hackernewsgithubprojects

📦 kunkka19xx/lgtm

lgtm: The Code Review Tool Built for AI Agents

lgtm is the terminal reviewer built specifically for the era of AI coding agents. The problem is simple: when an AI writes code faster than you can read, you cannot just scroll through a standard git diff and type out file paths to ask for changes. lgtm fixes this by living right next to your agent in a terminal pane. As the AI writes, the view updates instantly. The coolest part is that you can use vim motions to jump straight to a line, add a note, and send it back to the agent without ever retyping a file name. It even tracks where comments go when the code shifts.

🆔 @hackernewsgithubprojects

📦 browser-use/jev-ultrafast

Jev Ultrafast: The Browser Agent That Thinks Faster

Jev Ultrafast is the browser agent that decides its next move without guessing. It scans your screen to see every button and text box, then picks exactly what to click or type in a single step. Instead of looking at pictures of your screen, which is slow and blurry, it reads the actual structure of the page directly. This means it understands where things are instantly. The project proves this speed by searching for a flight from Zurich to London in just seven seconds. It types the city names and clicks the dates all on its own. Because it skips the visual guesswork, it works much faster than traditional tools.

📰 https://news.ycombinator.com/item?id=49735979

🆔 @hackernewsgithubprojects

📦 starnodes2024/comfyui-yue2-trainer

ComfyUI-YuE2-Trainer: Train Your Own Music Style

You can now teach a massive AI model to sound exactly like your specific songs without needing a computer science degree. This tool adds training nodes directly into ComfyUI, allowing you to feed it your own music files and a simple trigger word. Instead of writing code, you just drag in your audio, let it learn the specific style, instrumentation, and vocal timbre of your tracks, and it generates a ready-to-use adapter. This solves the problem of needing complex command-line tools to customize AI music generation. It is particularly useful for artists who want consistent sound profiles for their projects.

🆔 @hackernewsgithubprojects

📦 eval-exec/neomacs

Neomacs: The GPU-Powered Emacs Rewrite

Neomacs is the text editor rewrite that finally lets Emacs run on your graphics card. For forty years, Emacs relied on aging code that could not keep up with modern screens, so this project rebuilt the entire engine in Rust to handle rendering directly on your GPU. This means switching between documents or scrolling through files now happens at your monitor’s full refresh rate, making the interface feel alive rather than static. The best part is that it still runs your existing configuration exactly as before, so you get the speed of a game engine without losing your setup.

📰 https://news.ycombinator.com/item?id=49743386

🆔 @hackernewsgithubprojects

📦 mykeura/minimalist-themes-for-hermes

Minimalist Themes for Hermes Desktop: 18 Warm Palettes

The Hermes Desktop interface just got eighteen new, warm light color options that stick with you even when you switch between different work profiles. Most theme plugins force you to pick a new look every time you change your active profile, which is annoying if you like a consistent, readable workspace. This project fixes that by keeping your chosen minimalist palette selected automatically while you move between profiles. The palettes are intentionally light and warm, avoiding the usual dark mode fatigue, and they only affect the desktop interface, leaving your command line tools untouched.

🆔 @hackernewsgithubprojects

📦 itsjazii/pane

Pane: Track All Your AI Subscriptions in One Tray

Pane lets you check your remaining AI quota without opening a single browser tab. If you juggle subscriptions for tools like Claude, Codex, and Cursor, you know the frustration of logging into separate dashboards just to see if you have enough credit left for your next task. This free, open-source Windows app solves that by living quietly in your system tray. It automatically finds the login tokens your coding tools already have, queries each vendor directly for your real usage limits, and projects whether you will run out of credits before your weekly reset.

🆔 @hackernewsgithubprojects

📦 kubetail-org/kstack

kstack: AI for K8s

kstack is the AI skill pack that finally makes Kubernetes troubleshooting feel like chatting with a senior engineer. Instead of you manually typing out complex commands to hunt down errors, it lets you just describe the problem in plain English, like why a specific app is crashing or how much memory a service is using. It then quietly does the heavy lifting in the background, pulling logs and metrics so you get clear, useful answers without drowning in technical jargon or burning through your AI credits. If you work with cloud infrastructure, this is the shortcut that keeps you in control while doing the tedious detective work for you.

📰 https://news.ycombinator.com/item?id=48045711

🆔 @hackernewsgithubprojects

📦 rizqinrr/viserys-agent

Viserys Agent: The Workflow for AI Coders

Viserys Agent is the workflow pack that stops your AI coding assistant from guessing. It forces your AI agent to follow a strict six-step process, from defining the problem to shipping the code, instead of just spitting out whatever it thinks you want. The coolest part is that it includes specific skills like doubt-driven development, which tells the agent to actively challenge its own assumptions before writing a single line of code. This turns a chatbot into a disciplined engineer that actually checks its work. If you are using AI to write code, this is the layer that keeps it from hallucinating its way into a mess.

🆔 @hackernewsgithubprojects

📦 grafana/agento11y

Grafana Agento11y: See Inside Your AI Agents

Grafana Agento11y is the observability suite that finally lets you see exactly what your AI coding agents are doing. It wraps tools like Cursor, Claude Code, and Codex to capture every prompt, tool call, and token usage, sending that data to Grafana Cloud for analysis. You get a unified dashboard showing costs, performance, and privacy settings across all your different agents without manual logging. This helps developers spot expensive loops, debug failures, and track how models behave over time. If you are building or using AI agents, this project gives you the clarity to understand and manage them effectively.

🆔 @hackernewsgithubprojects

📦 monid-ai/monid

Monid: The Universal Router for AI Agents

One API key, one address, and your AI agent can reach over two thousand distinct tools instantly. That is the core promise of Monid, a platform that acts like a universal router for agent capabilities. It solves the messy problem of integrating dozens of different vendors by wrapping them in a single, standardized interface. Instead of your agent fumbling through specific documentation for every new service, it simply asks Monid to find the right tool. The system even ranks options by cost and speed at the moment of the request. This means your agent always picks the most efficient and cheapest path to get the job done.

📰 https://news.ycombinator.com/item?id=49730644

🆔 @hackernewsgithubprojects

📦 declare-lab/della

DELLA: Smarter Model Merging

DELLA is the model merging technique that finally stops AI models from fighting each other. When researchers combine multiple specialized AI models into one, their internal settings often clash, causing performance to drop. This project solves that problem by ranking every single setting based on its size and then randomly discarding the smallest, least important ones. It then mathematically boosts the remaining strong settings to fill the gap. This simple idea prevents the different models from interfering with each other, resulting in a combined system that handles language, math, and code better than any previous merging method.

🆔 @hackernewsgithubprojects

📦 rerun-io/examples-monorepo

One Home for Every Vision Project

Rerun’s examples monorepo packs over twenty distinct computer vision applications into one unified workspace, so you can test everything from hand tracking to 3D scene reconstruction without juggling a dozen different setups. It solves the classic dependency mess by using a single configuration file to manage every environment, meaning your camera feeds and neural networks run exactly the same way on your laptop as they do on a server. You get instant access to tools that turn raw video into interactive 3D models, all backed by one consistent build system.

🆔 @hackernewsgithubprojects

📦 venkatapgummadi/ascend

ASCEND: Catches AI Coding Mistakes

ASCEND stops the bugs your AI coding assistant introduces before they reach production. While most pipelines just run tests, this tool adds a strict security gate that actually blocks bad code from moving forward. The standout feature is how it handles AI drift. When your helper changes something unexpectedly, ASCEND compares the code structure to catch errors that normal tests miss. It then uses AI to fix the conflict automatically. This keeps your codebase safe without slowing down your team. It is a smart way to balance speed with security. If you use AI to write code, you need this safety net.

📰 https://news.ycombinator.com/item?id=47858020

🆔 @hackernewsgithubprojects

📦 doelfke/ninfer-yarn

Run Qwen 3.8 B on One GPU

Ninfer Yarn is the inference engine that squeezes a massive language model onto a single consumer graphics card. It runs the twenty-seven billion parameter Qwen model with a huge context window on an RTX 5090, which is wild because that usually requires expensive server racks. The project focuses on raw speed by using specialized math kernels and efficient memory handling, so you get fast responses without needing a data center. It is built for people who want high-performance AI locally without the massive power bill or hardware cost. If you are curious about running big models at home, this is the project to watch.

🆔 @hackernewsgithubprojects

📦 summerengine/summer

Summer: The Game Engine for AI Agents

Build playable games using your favorite AI coding assistant by giving it direct control over a real game engine. Summer is an open source package that turns your text chat agent into a capable game developer. It connects your AI directly to the engine so it can build scenes, write logic, and test the results in real time. The coolest part is that the agent can actually play the game and read the debugger to fix its own mistakes. It also remembers your project details so you never have to repeat yourself. If you want to make games with AI, this is the tool you need.

🆔 @hackernewsgithubprojects

📦 gargpratyush/jev-router

Jev Router: Save Money on AI Coding

Route your coding tasks to the cheapest AI model that can actually handle them. Jev Router wraps your favorite command-line tools, automatically deciding between fast, balanced, and strong models for every single request. Instead of paying top dollar for complex reasoning on simple code fixes, it keeps the interface you already know while quietly swapping out the engine behind the scenes. This means you get premium results for heavy lifting and snappy responses for quick fixes, all without changing how you work. It is a smart way to cut costs without sacrificing the quality of your code.

📰 https://news.ycombinator.com/item?id=49746321

🆔 @hackernewsgithubprojects

📦 tinyhumansai/opencompany

opencompany: The AI Office Simulation

opencompany is the agent framework that simulates an entire workforce working toward a single goal. It solves the problem of scattered AI helpers by organizing them into distinct, professional roles, such as accountants, designers, or researchers, each with specific duties and shared company standards. You define a business unit, like a software startup or a law firm, and the system assigns agents to their jobs, allowing them to collaborate on real tasks like drafting contracts or closing quarterly books. This turns a loose collection of prompts into a coordinated team effort. Instead of asking a single AI to do everything, you watch a digital company operate with structure and intent.

🆔 @hackernewsgithubprojects

📦 lovablelabs/oj

OJ: The Rust React Build Tool That Fixes Memory

OJ is the build tool that finally makes running React apps on a tight budget a reality. It is a native Rust compiler that replaces the heavy JavaScript overhead of traditional development environments, meaning it starts up faster and uses a tiny fraction of the computer memory you are used to wasting. While other tools can eat up gigabytes of your system’s resources, OJ keeps things lean and snappy, even when you are running multiple builds at once for testing or deployment. This makes it a game changer for anyone who cares about efficiency, proving that you do not have to sacrifice speed for reliability.

📰 https://news.ycombinator.com/item?id=49746713

🆔 @hackernewsgithubprojects

📦 syv-ai/hyperqwen

HyperQwen: 27B on 24GB

HyperQwen lets a single twenty-four gigabyte graphics card serve a massive twenty-seven billion parameter language model at speeds that rival dedicated server clusters. This project solves a common problem where huge AI models simply do not fit on consumer hardware by using a specialized pipeline to compress and optimize the data flow. It achieves this through a series of precise technical patches and a unique way of predicting the next words to generate, which dramatically speeds up the process. The result is a fast and responsive local assistant that runs entirely on your own machine.

🆔 @hackernewsgithubprojects

📦 raemote/raemote_server

Raemote: Your Home Server Pocket Edition

Raemote is the private bridge that puts your home server apps in your pocket. If you run media players or smart home dashboards on your computer, you usually have to be in the same room or set up complicated network tricks to reach them from your phone. Raemote solves this by running quietly in the background, finding the web apps already running on your machine, and creating a secure, encrypted tunnel directly to your iOS device. No accounts, no public IP addresses, and no port forwarding required. It works like a magic remote control for your own hardware, whether you are on the same Wi-Fi or miles away.

🆔 @hackernewsgithubprojects