Trending & Rising

High-velocity repos and community favorites.

openclaw/openclaw+1242 stars/day
mattpocock/skills+1132 stars/day
affaan-m/ECC+1050 stars/day
obra/superpowers+818 stars/day
NousResearch/hermes-agent+572 stars/day

Trending Now

High-velocity repositories gaining rapid community traction.

MakcReMA

MakcRe/KuGouMusicApi

酷狗音乐 Node.js API service

1,000
587
empiraEM

empira/PDFsharp

PDFsharp and MigraDoc Foundation for .NET 6 and .NET Framework

1,000
261
zycgitZY

zycgit/hasor-old

Hasor是一套基于 Java 语言的开发框架,区别于其它框架的是 Hasor 有着自己一套完整的体系,同时还可以和先有技术体系做到完美融合。它包含:IoC/Aop容器框架、Web框架、Jdbc框架、RSF分布式RPC框架、DataQL引擎,等几块。

1,000
267

Hidden Gems

High-utility projects flying under the radar. Low stars, high potential.

NEWdeepseek-ai/deepseek-harness1 month old
NEWbilawalsidhu/gods-eye-view3 months old
NEWDietrichGebert/ponytail3 months old
NEWodysseus-dev/odysseus4 months old
NEWayghri/i-have-adhd4 months old

New Arrivals

Recently created projects with active commits and initial traction.

AstraZenecaAS

AstraZeneca/DiffAbXL

The official implementation of DiffAbXL benchmarked in the paper "Exploring Log-Likelihood Scores for Ranking Antibody Sequence Designs", formerly titled "Benchmarking Generative Models for Antibody Design".

100
9
laduniestuLA

laduniestu/better-next

Nextjs starterkit using Better Auth

100
4
dcodesdevDC

dcodesdev/LetterSpace

Self-hosted open source newsletter platform for managing subscribers and sending campaigns

100
2

Early Stage

Young, active projects building towards their first 100 stars.

Latest Picks

Handpicked repositories selected for utility and relevance.

ECC interface preview

ECC: Memory System for Claude Code That Survives Sessions

Claude Code and similar AI assistants lose all context when you close the terminal. ECC solves this through persistent memory hooks, coordinated planning workflows, and reusable skills—letting your AI assistant pick up exactly where it left off. Built by developers who used it to win Anthropic's hackathon with Zenith.chat.

affaan-mAF

ECC

affaan-m

skills interface preview

Matt Pocock's Skills: Teaching AI Agents to Debug Themselves

AI coding assistants often misunderstand intent, producing code that compiles but misses the mark. Matt Pocock's skills repository formalizes engineering discipline—static types, TDD, deliberate debugging—into feedback loops that help agents course-correct. We examine how 274k stars reflect demand for structured workflows, address controversies around skill bloat and secret exposure, and compare complementary approaches like Obra's Superpowers.

mattpocockMA

skills

mattpocock

omarchy interface preview

Omarchy: The Tiling WM Distro Without Assembly Tax

Tiling window managers like Hyprland offer power and aesthetics, but users had to assemble and maintain the desktop environment themselves. Omarchy packages the entire stack—WM, shell, themes, system agents—into a distribution with stability guarantees. 37signals' adoption and 200K+ downloads in 18 days show this middle ground between Ubuntu and Arch+Hyprland filled a legitimate need.

omacomOM

omarchy

omacom

CL4R1T4S interface preview

CL4R1T4S: The Hidden System Prompts Behind Every AI

Every AI assistant operates under hidden instructions users never see. CL4R1T4S documents these system prompts across dozens of AI products, revealing the guardrails, biases, and capabilities baked into each system. The archive addresses a fundamental transparency gap: users interact with AI daily without understanding what shapes its responses.

elder-pliniusEL

CL4R1T4S

elder-plinius

ai-job-search interface preview

The Job Search Tool That Stops Before Clicking Submit

Professionals face a false choice: spend hours tailoring every application by hand, or let bots spray-and-pray on your behalf. The ai-job-search repository runs the entire pipeline—scraping, fit analysis, CV customization—but deliberately stops before final submission. You review, you decide, you click send.

MadsLorentzenMA

ai-job-search

MadsLorentzen

codegraph interface preview

CodeGraph: Stop AI Agents From Re-Reading Your Codebase

AI coding agents waste context windows playing detective—reading files repeatedly to reconstruct call graphs and dependencies. CodeGraph solves this by building a persistent knowledge graph that answers questions about your codebase without the grep/glob/re-read cycle that burns thousands of tokens per query.

colbymchenryCO

codegraph

colbymchenry