📦 ameyanagi/reshiki

ReShiki: The Rust-Based Molecular Drawing Workspace

ReShiki is a standalone desktop application that lets chemists draw complex molecules and reaction schemes without needing to install Python or RDKit, handling all chemical logic entirely in Rust. It functions as a complete workspace where you can sketch structures, apply professional styling, and export high-quality figures directly from your computer. The real surprise is that it runs completely offline, meaning your data stays private while you work. This tool bridges the gap between quick sketching and publication-ready figures, making it a fascinating option for developers interested in high-performance scientific software.

🆔 @hackernewsgithubprojects

📦 harbor-framework/terminal-bench-2-1

Terminal Bench 2.1

Terminal Bench 2.1 is the benchmark suite that actually tests if your A.I. can finish hard jobs. It gives agents real container environments to fix bugs, assemble proteins, or patch security holes. This update fixes twenty-six previous tasks to stop agents from cheating to get high scores. It matters because we need to know if A.I. can handle messy, real-world work. If you care about A.I. that delivers results, this is the standard to watch. It is the honest test for the future of digital workers. See you in the next one.

🆔 @hackernewsgithubprojects

📦 liquid4all/liquid-audio

Liquid Audio: Real-Time Speech AI

Liquid Audio is the speech foundation model that finally lets AI talk to you in real time without lag. Built by Liquid AI, it handles the entire loop, hearing you and answering with natural voice instead of just reading text. The coolest part is how it processes sound and language together, so it can hold a conversation while still understanding exactly what you meant. It is lightweight enough to run on regular hardware, making interactive voice agents actually feasible for developers. If you have been waiting for an open model that handles audio like a first-class citizen, this is the one to check out.

🆔 @hackernewsgithubprojects

📦 clerk/skills

Clerk Skills

Clerk Skills is the cheat code that turns your AI coding assistant into a Clerk expert. It stops the model from guessing how to handle authentication by giving it a complete, organized library of proven patterns. The coolest part is that it actually thinks. When you ask your agent to add login to your Vue or Next dot JS app, a router skill instantly picks the right guide from the bunch. It knows exactly which hooks to use or how to protect your server routes. No more messy, half-finished code. You just get clean, production-ready authentication logic that fits your specific framework perfectly.

🆔 @hackernewsgithubprojects

📦 getsentry/skills

Sentry Skills: The Open Source Playbook for AI Coding

Sentry has published the exact instruction sets their engineers use to program their AI coding assistants. This repository, called Skills, acts as a shared memory bank that teaches artificial intelligence how to think like a Sentry developer. It covers everything from writing clean code to navigating their specific security standards. The most surprising part is that these are not just generic tips; they are precise, tested workflows that ensure every line of code generated matches the company's highest standards. By opening up these internal tools, Sentry gives every developer a free, professional-grade cheat sheet for training their own AI helpers.

🆔 @hackernewsgithubprojects

📦 mastra-ai/skills

Skills: The Knowledge Layer for Mastra

Skills is the knowledge layer that fixes outdated coding agent memory for the Mastra framework. You know how AI assistants often write code that looks right but fails because they rely on old training data? This repository solves that by giving agents a fresh, verified map of current APIs and workflows. Instead of guessing, the agent checks live documentation directly from your installed packages or remote sources before writing a single line of code. It ensures your AI assistant actually understands your specific version of Mastra, preventing frustrating errors and broken builds. This means you get accurate, production-ready code every time, without manually fact-checking the AI’s every suggestion.

🆔 @hackernewsgithubprojects

📦 mcollina/skills

Skills: Teaching AI to Code Node.js

Matteo Collina, the lead maintainer of the Node.js project, has published a public library of instructions called Skills that significantly boosts the accuracy of AI coding assistants. The repository contains carefully crafted text files that teach large language models specific best practices for building fast and secure Node.js applications. When you feed these instructions to an AI, the model stops guessing and starts writing code that follows professional standards. This is interesting because it shows that we can improve AI output simply by providing better context and clear guidelines. It turns a generic chatbot into a specialized helper that understands the nuances of server development.

🆔 @hackernewsgithubprojects

📦 neondatabase/agent-skills

Agent Skills: How AI Writes Better Database Code

Your AI coding assistant can now figure out why your database bill is suddenly expensive. The project called Agent Skills is a collection of instruction sets that teach artificial intelligence how to work with the Neon cloud database platform. Instead of guessing how to write queries, the agent reads these detailed guides to handle everything from setting up authentication to optimizing data transfer. This solves the problem of generic AI producing inefficient or incorrect code for complex backend systems. By giving the machine specific knowledge about how Neon works, it can actually fix your code and save you money.

🆔 @hackernewsgithubprojects

📦 posthog/skills

PostHog Skills: The Missing Instruction Manual for Your Data

PostHog Skills is the instruction manual that turns an AI assistant into a data expert for your product. Instead of guessing how to write queries or interpret user behavior, you now have a massive library of specific, ready-made guides for every task imaginable. It solves the problem of AI hallucinations by giving it precise, vetted steps for everything from fixing broken dashboards to tracking exactly how much your AI models cost. This is interesting because it removes the guesswork from data analysis, making sure your insights are always grounded in reality.

🆔 @hackernewsgithubprojects

📦 prisma/skills

Prisma Skills Explained

Prisma Skills is the instruction library that teaches AI coding agents how to handle databases without breaking your data. It solves the problem of unpredictable automation by packaging specific, verified guides for tasks like running migrations, setting up new clients, and upgrading complex version changes. When you ask an assistant to update your schema, the agent loads these packaged rules to ensure it follows the exact command syntax and safety protocols required for your specific database provider. This means your AI helper stops guessing at terminal commands and starts executing precise, safe database operations.

🆔 @hackernewsgithubprojects

📦 semgrep/skills

Skills: AI Coding Security

Skills is the agent toolkit that turns your AI coder into a security expert. Semgrep packaged thousands of real-world vulnerability patterns into simple instructions your coding assistant reads automatically. The standout feature is how it handles AI applications; it teaches your bot to spot prompt injection and data poisoning attacks before they happen. Instead of relying on guesswork, the agent checks specific rules for every line of code you write. It covers everything from database queries to cloud settings, ensuring your AI never suggests unsafe shortcuts. You install it once, and your helper instantly knows how to keep your software safe without you needing to memorize complex security standards.

🆔 @hackernewsgithubprojects

📦 0xkuch/fibo

Fibo: The Truth About Your Estimates

Your code history knows exactly how much your time estimates lie. A new tool called Fibo acts like a blame report for your promises, comparing what you said a task would take against the real hours you spent coding it. It calculates your personal multiplier, revealing that while you think you need two days, you actually take nearly three times that long. By pulling data straight from your Git history and trackers like Jira, it shows you exactly where your intuition fails, without needing a database or a server.

🆔 @hackernewsgithubprojects

📦 alex314618-create/jevrev

JevRev: The Decision Layer for Your AI

JevRev stops your coding agent from burning cash on bad ideas. It acts as a strict filter between your LLM and the final code, forcing the AI to prove its choices before moving forward. The coolest part is how it handles a simple task like speeding up a data parser. The AI suggests a quick shortcut that looks perfect on paper, but JevRev runs a real test and catches the hidden error immediately. Instead of blindly trusting the first answer, it audits the evidence and picks the slower, reliable method that actually works. It turns guesswork into a verified process you can inspect.

🆔 @hackernewsgithubprojects

📦 resend/resend-skills

Resend Skills: Give Your AI an Inbox

Teach your coding assistant to actually send and receive email for you. This collection hands your AI agent a secure inbox so it can reply to messages without you manually wiring up every server. The coolest part? It builds in safety rails that stop random strangers from using your bot to execute shady commands. It handles the messy background work of verifying who is sending an email before your AI even sees it. No more guessing if a message is legit; your agent just knows. If you are building smart assistants, this is the missing piece that makes email conversations feel real and safe.

🆔 @hackernewsgithubprojects

📦 amitshekhariitbhu/ai-engineer-roadmap

The AI Engineer Roadmap

One person mapped out the entire journey to becoming an AI engineer, breaking down eighteen distinct steps with a free blog for every single concept. It is not just a list of tools, but a structured path that starts from basic machine learning foundations and walks you through the inner workings of transformers, how models actually generate text, and how to build reliable agents. The real value here is the plain language. Instead of drowning you in academic papers, it explains complex ideas like attention mechanisms and vector search in simple terms anyone can follow.

🆔 @hackernewsgithubprojects

📦 amitshekhariitbhu/ai-engineering-course

The Complete AI Engineering Path

One repository now maps the entire journey from basic machine learning to complex AI system design, a rare feat for a free resource. This course guides developers through neural networks, transformers, and large language models, covering practical steps like fine-tuning and building AI agents. It also tackles critical areas often ignored, such as model evaluation and AI safety, ensuring you understand not just how to build these systems, but how to maintain them responsibly. The material is structured to take you from fundamental concepts to advanced engineering challenges, making it a comprehensive roadmap for anyone serious about mastering modern artificial intelligence.

🆔 @hackernewsgithubprojects

📦 anishfn/shapeshift

Shapeshift: The Input That Becomes What You Mean

Shapeshift lets you type a single sentence and watch a plain text box instantly morph into the exact tool you need, turning your raw words into a working calendar invite, bill splitter, or countdown timer. It solves the annoying habit of opening five different apps just to handle quick daily tasks, all from one smart input field. The coolest part is that it works entirely offline using a built-in keyword classifier, so you get that instant, magical morphing effect without needing an internet connection or a heavy AI backend. If you love seeing text turn into useful, interactive widgets on the fly, this is the project to check out.

🆔 @hackernewsgithubprojects

📦 ashleydarosa/cc-checker-validator-generator

cc-checker-validator-generator

This command line tool lets you instantly generate and test mock payment card numbers that bypass basic frontend validation checks without hitting the live network. By mimicking the math behind real payment networks, the software builds perfectly structured test numbers that pass the industry standard Luhn checksum algorithm out of the box. It is built for developers who need to run rigorous checkout forms through their paces across major card brands using completely offline, custom generation templates. You can easily test edge cases, look up issuer bank identification numbers, and export bulk batches to format files like comma separated values.

🆔 @hackernewsgithubprojects

📦 asokurasu/text-humanizer

text-humanizer

You can now automatically strip away the mechanical, robotic feel of computer-generated writing without paying a dime. The project text-humanizer is a clever open-source tool that routes your text through a multi-step linguistic pipeline to make it sound completely natural. Instead of just swapping words, it translates your content across multiple languages—like Turkish, Japanese, and Chinese—using translation engines like DeepL and Google, before reconstructing it back into your original language using a DeepSeek language model. This process naturally distorts rigid sentence structures and eliminates repetitive patterns, leaving you with smooth, human-like writing that easily slips past most automated detectors.

🆔 @hackernewsgithubprojects

📦 caxete/crypto-tax-calculator

Crypto Tax Calculator: The All-in-One Tax Solver

This open-source tool ingests raw trade data from over thirty exchanges and wallets, automatically sorting your complex history into a single, audit-ready tax report. It solves the nightmare of manually tracking every staking reward, airdrop, and swap by applying precise tax rules to your specific jurisdiction. You simply drop in your CSV file, and the engine handles the rest, fetching historical prices to calculate your exact capital gains. It is built for transparency, ensuring you pay the right amount without the headache. If you hold crypto, this is your shortcut to compliance. Stop guessing and start calculating.

🆔 @hackernewsgithubprojects

📦 bridge-mind/bridgeclip

BridgeClip: The Open-Source Clipping App

BridgeClip turns massive video files into captioned shorts directly on your computer. It finds the best moments using AI, cuts them into the perfect aspect ratio, and burns in stylish word-by-word text. The coolest part is that everything happens locally. Your data, your keys, your machine. No middleman servers. Just a clean, open-source tool that respects your privacy while doing the heavy lifting for you. It is the smart, private way to handle your content workflow. If you create long-form video, this is the missing piece in your desktop setup. Stop letting cloud services hold your footage hostage. Get control back with BridgeClip.

🆔 @hackernewsgithubprojects