Most starred

PasLLM - LLM inference engine in Object Pascal (synced from my private work repository)

Observe ROS 2 DDS and Zenoh traffic, inspect graphs and topic metrics, and record messages with a CLI or desktop GUI.

Open high-speed micro quadcopter R&D platform with printable airframe references and reproducible Blackbox flight tests.

OpenAI Privacy hackathon - Paris, 🏆 winning project

Ethiack AI Pentesting Benchmark Framework

Real machine coding & low-level design (LLD) interview questions asked at Amazon, Microsoft, Uber, Flipkart & more — mapped to design patterns, with reference solutions in 5 languages.

Switch AI coding tools between your accounts. App and command line for macOS, Windows and Linux.

Self-hostable RAG backend for document knowledge: hybrid search, cited answers, and agentic deep reading. Java · Spring Boot · Spring AI · Elasticsearch.

麦当劳 MCP 技能：待评价清单、好评兑礼进度、发票报销门槛凑单、单次抽奖确认。99 项测试，非官方产品。

Live-preview Ghostty terminal configuration for theme, fonts and UI.

A versatile WiFi auditing swiss-knife for ESP32. Provides a professional C++ API for network discovery, performance testing, and security auditing.

An opinionated list of awesome Pydantic-AI frameworks, libraries, software and resources.

A comprehensive GraphRAG (Graph Retrieval-Augmented Generation) system designed for biomedical research

GibRAM is an in-memory knowledge graph server designed for retrieval augmented generation (RAG / GraphRAG) workflows.

AI-powered media search — find images and videos using natural language or visual queries

600 KB WASM runtime for Cactus Compute's Needle AI tool-calling models, Needle 3, 2 and 1 from one build. Browser, Cloudflare Workers, Node.js, Python, C FFI and no_std. Token-exact with the JAX reference. No backend, no API key.

WAVE-Go: Go2-W world-model action adaptation, verified mapless charging, and Cosmos3-Edge hybrid long-range navigation demos

ZKE（Z Kubernetes Engine）：AI 原生的 Kubernetes 云操作环境，桌面式多集群控制台加受控 AIOps Agent，基于 Server + Agent 与 QUIC/mTLS，适用于私有云、混合云及边缘环境

Self-hosted LLM gateway that routes requests across AI providers (OpenAI, Anthropic, Gemini, Mistral, Ollama) using intelligent multi-policy scoring — including an LLM-native routing policy. Drop-in compatible: just swap the base URL. No database required, built-in cost tracking, budget enforcement and multi-tenant isolation.

AI guardian for manual crypto traders — risk monitoring, strategy validation & emotional trading detection. No trade execution.

View SQLite databases

ИИкона — AI-коннектор для 1С:Предприятие 8.3: LLM-провайдеры, агентская петля (function calling), RAG, MCP-сервер, мониторинг ошибок с ИИ-анализом, генератор диаграмм

Data streaming runtime focused on performance, consistency, and extensibility. Write plugins in Rust or WASM and process data with data guarantees.

Official codebase for Unitas: A 3D-Native World Action Model for Embodied Manipulation

Sandbox with hardware IO and hot reload for ESP32 devices

A senior security engineer's actual browser bookmarks. 40+ curated cybersecurity resources for recon, OSINT, exploitation, privilege escalation, and malware analysis.

帮你点麦当劳最省的 Skill：自动叠加优惠券、积分和门店活动，实时核价算出最省付法｜基于麦当劳中国 MCP

Slidev-style presentations written in markdown, compiled with Angular. The core app and the slide-ng command line.

从零开始，看懂 DeepSeek Harness 的「万物皆可插件」— 零基础插件开发教程（含 22 个中文技能实例）| Zero-to-plugin tutorial for DeepSeek Harness

POOM is an open-source multitool platform built to Pentest. Play. Create. It is designed for makers, security learners, gamers, and engineers who want a compact device capable of real embedded workflows across wireless security, data capture, automation, and interaction.

The LuxAlgo ecosystem as MCP tools: free and keyless.

Physical AI skills for coding agents, with field-tested robotics guidance, the robium-ai CLI, and a learning engine.

Autonomous backend platform. Your coding agent builds it over MCP, Backenly keeps it running: Postgres, REST APIs, auth, storage, realtime, and functions, with every change governed, verified, and reversible.

Claude Code plugin: describe an object or show a photo, and Claude builds a clean 3D model in Blender, for games, the web, AR or 3D printing. It animates characters too. New in 3.0: Kiln Studio, a local page to view, check and fix your models with Claude.

High-throughput synthetic and pretraining dataset sifter for TypeSafe Jev. Rust streaming core, Parquet and JSONL I/O, typed Choice/Score/Noul judgments, speculative fan-out, 24.0 rows/sec measured.

One MCP server, two stores: App Store Connect + Google Play.

ROS Noetic 两轮差速自主移动机器人：ESP32 底盘(PID/编码器/IMU) + J1900 车载(LiDAR/相机) + PC(SLAM/导航/人体跟随)，三机分布式全链路开源实现

Your AI usage, quota, and balance — as a desk pet. 🐾 A local companion service that normalizes every AI provider into one LAN feed, plus ESP32-S3 firmware for the device that shows it.

Open-source virtual pet firmware for the M5Stack Cardputer.

📈 B2B SaaS Growth Playbook — PLG & SLG strategies from PMF to $10M ARR. Real case studies: HeyGen, Deel, Vercel, Supabase. Affiliate programs, channel partnerships, enterprise sales.

Best Claude Opus 5.5 video & visual demos, curated with prompts · 149 精选可视化案例 + 1019 完整清单（游戏/3D/动效/讲解/可视化）

Kinetiq Gym: a programmable smart gym built as open hardware, in development. Website, build preview, docs and priced parts lists. CAD, build files and software are released after the prototype is tested.

AI-powered XAUUSD (Gold) trading bot with XGBoost ML, Smart Money Concepts (SMC), and HMM regime detection for MetaTrader 5

KDE Plasma for Windows — step-by-step install and first-run setup on Windows 10 and 11, with download steps in the README.

PROMETHEVS: run the full DeepSeek-V4.1-Flash-1017 (510 GB) on a gaming PC with 64 GB RAM and a 12 GB GPU. PROMETHEVS engine (compress-then-correct), OpenAI-compatible local server, report and logs.

An open-source local print agent that enables silent printing from web applications.一个开源本地打印代理，让 Web 应用实现稳定的静默打印。

Selfhost modern LLM stacks. Run the whole fleet from your terminal

TypeSafe Jev for DeepSeek Harness, the Model Context Protocol, and plain Node: typed judgments instead of prose, offline by default.

High-fidelity process tracking, deterministic replay, and AI-powered debugging for Laravel

闲鱼多店铺智能运营系统，集成 AI 客服、自动发货、商品知识库与智能运维。