mvanhorn/last30days-skill

AI agent skill that researches any topic across Reddit, X, YouTube, HN, Polymarket, and the web - then synthesizes a grounded summary

View on GitHub

What it does

Last30Days is a plug-in skill for the Claude AI coding assistant that automatically researches any topic across Reddit, X, YouTube, Hacker News, Polymarket, and Bluesky, then produces a cited summary of what people are actually talking about right now. Think of it as a one-command briefing tool that scans the social web for the past 30 days and distills the signal into a readable report, saved automatically to your computer.

Why it matters

As AI tools and markets shift weekly, founders and product teams who can quickly validate what's gaining traction — before it becomes mainstream knowledge — have a real edge in prioritization and positioning. The 15,000+ stars suggest strong demand for ambient, automated trend intelligence baked directly into developer workflows rather than requiring separate research tools.

Why it's trending

The idea of a single command that scans Reddit, X, YouTube, Hacker News, and more to produce a cited briefing on any topic clearly struck a nerve — this week the project went from 337 new stars to nearly 10,000, a 2,866% jump that signals viral sharing rather than steady organic growth. With 106 commits in the past 30 days and 720 new forks this week alone, builders aren't just starring it, they're actively pulling it down and building with it. That said, the near-perfect star velocity combined with a tiny contributor base of just 9 people warrants a closer look before betting heavily on it — momentum this sudden can reflect a coordinated push as much as genuine community adoption.

57Hot

Gaining traction — heating up

Stars

25.7k

Forks

2.2k

Contributors

Language

Python

Related projects

opencv/opencv

64Hot

OpenCV is a free, widely-used software library that gives computers the ability to interpret and understand visual information — like recognizing faces, reading text from images, or detecting objects in video. It handles everything from basic photo editing tasks to powering sophisticated AI-driven vision systems used in robotics, self-driving cars, and medical imaging.

// why it matters With 87,000+ stars and over 2,400 contributors, OpenCV is effectively the industry-standard foundation for any product that needs to 'see' — meaning builders can skip years of foundational work and ship vision-powered features faster. For founders and investors, this signals that computer vision capabilities are now a commodity layer, shifting competitive advantage toward the application and data layer rather than the underlying vision technology.

C++87.5k stars56.6k forks2414 contrib

AI / Machine Learning GitHub ↗

ROCm/aiter

60Hot

AITER is AMD's open-source library of high-performance building blocks that make AI models run faster on AMD hardware, supporting everything from basic AI operations to complex training and multi-GPU coordination. Think of it as a toolbox that lets AI software teams tap into AMD's chip capabilities without having to write low-level hardware code themselves.

// why it matters As AI infrastructure costs soar, builders are actively exploring alternatives to Nvidia's dominant GPU ecosystem, and AMD is positioning AITER as the key compatibility layer that makes switching or diversifying hardware more practical. For founders and PMs building AI products, this means AMD GPUs become a more credible option for cost reduction or supply chain diversification — especially relevant as demand for AI compute continues to outpace supply.

Python428 stars307 forks200 contrib

AI / Machine Learning GitHub ↗

scikit-learn/scikit-learn

57Hot

Scikit-learn is a free Python library that gives developers ready-made tools for building machine learning models — the algorithms that let software learn patterns from data to make predictions, classifications, and decisions. It covers everything from detecting spam to predicting customer churn, all without requiring deep expertise in the underlying math.

// why it matters With over 65,000 stars and nearly 3,500 contributors, scikit-learn is effectively the industry standard starting point for adding AI-driven features to products, meaning most data science talent already knows it and most ML workflows are built on top of it. For founders and PMs, it dramatically lowers the cost and time to prototype intelligent features, making it a critical dependency to understand when evaluating build-vs-buy decisions around AI capabilities.

Python66.1k stars27.0k forks3496 contrib

AI / Machine Learning GitHub ↗

mindsdb/mindsdb

57Hot

MindsDB lets AI agents securely search and retrieve information from virtually any database or data source using plain language queries, acting as a smart middleman between AI systems and company data. It's paired with ANTON, a personal AI agent that can take that data access and turn it into real actions — like sending emails, building dashboards, or calling external services — without requiring any technical setup.

// why it matters As businesses race to build AI-powered products, the hardest problem isn't the AI itself — it's safely connecting AI to the messy, siloed data that already exists inside companies, and MindsDB with nearly 40,000 GitHub stars signals strong developer validation for exactly that problem. Founders and investors should pay attention because whoever controls the layer between AI agents and enterprise data is positioned to become critical infrastructure for the next wave of business software.

Python39.2k stars6.2k forks888 contrib

AI / Machine Learning GitHub ↗

// SUBSCRIBE

The repos that moved this week, why they matter, and what to watch next. One email. No noise.