Havuz

250+ real-world TypeScript AI projects: workflows, agents, and multi-agent systems with production-ready architecture, not chatbot demos.

Open-source AI that watches and hears your screen and reacts live as any personality you design — for faceless content, autonomous AI stream

The Karpathy treatment for OpenClaw - stripped to ε. 515 lines of Python, 6 files, one dependency.

Spec-Driven Development framework that enables humans and AI agents to build better software by giving them small, local instructions right

AI-native publishing system — build books like software. 6-phase pipeline from blank page to EPUB/PDF, with 16 skills, 7 agents, and 8 genre

Video2GUI: Synthesizing Large-Scale Interaction Trajectories for Generalized GUI Agent Pretraining (ICML2026)

Developer-friendly LLM agent harness built on Jido and Jido AI

Local streaming reverse proxy between AI coding agents (Claude Code, Cursor, Codex) and model APIs (Anthropic, OpenAI, Gemini, MiniMax). Met

国内公募基金数据采集与AI辅助分析工具 — 全栈FastAPI+React，集成AKShare多源数据、风险评分引擎、回测验证、LLM可选分析

Longitudinal relational modeling for reflective agents: behavioral signals, evolving subject profiles, inspectable runtime governance.

Unified native desktop UI for Claude Code, Gemini CLI, Hermes, Crush, Codex, and Cursor — multi-agent, MCP-native, written in Rust + GPUI.

Autonomous AI-powered penetration testing platform. LLM-driven recon, vulnerability analysis, and exploit validation for internal & external

Apache Camel-style integration framework for .NET. 30+ EIP patterns, 29 transports (Kafka, RabbitMQ, MQTT, S3, gRPC, SFTP, SQL...), fluent C

TurboQuant MLX: Flagship high-performance model quantization for Apple Silicon. Squeeze maximum inference speed and memory efficiency out of

The most powerful open-source TradingView AI agent. 50 tools, 15 models (Claude, GPT, Gemini, Qwen, DeepSeek, MiniMax), Pine Script v6, quan

LLM spend tracking and budgets for Rails — by user, feature, or any tag, in your own database, no proxy.

Fastest way to scaffold your LLM powered personal knowledge base - Obsidian, Claude Code, qmd, Github

Open source GitLab MCP server for AI assistants: 2-tool dynamic find/execute over 850+ GitLab actions (1,000+ Enterprise), stdio/HTTP/OAuth,

Self-hosted AI security proxy. Redact PII, block prompt injection, route to any LLM provider. OpenAI-compatible.

GPT-5 / Claude Opus 4.7 / Gemini 3.1 / Sora 2 / Suno / DeepSeek / Kimi - AI API integration cookbook for Cursor, Cline, Claude Code, ChatBox

Convert long AI conversations into portable conversation state graphs for LLM handoffs.

Turn your AI coding agent into a tutor. It learns your level and adapts to you

Curion is a project-local memory layer for AI coding agents, published as an MCP server for Claude Code, Codex, OpenCode, and other MCP clie

Canonical eight invariants for AI-assisted engineering: README SSOT, skill guidance, and lightweight checks.

Fully local: chat, a coding agent, and a design studio in one private desktop app.

ONE API，LLM_Proxy，统一代理，支持请求响应自定义转换。

QodeX — a local-first LLM agent & AI coding CLI agent for your terminal. Runs local models (Qwen3-Coder via Ollama/LM Studio) with determini

Turn AI coding CLIs (Claude Code, Codex, Antigravity) into AI employees on LINE, Telegram, Discord, Slack and 7 more chat apps. A judge revi

AI terminal for coding agents: run Claude Code, Codex, and shells side by side on your computer or remote servers. Agents ask each other for

Android AI agent app. Kotlin + Rust via UniFFI.

Self-hosted OpenAI-compatible API proxy for Google Gemini — authenticated via OAuth 2.0 + PKCE, no paid API key required. Built with Bun + H

"Does this actually work with my agent?" An open, versioned registry of AI-agent artifacts: skills, plugins, MCP servers, models, etc. with

Open-source Claude Code plugin porting the internal /skillify skill: turn a finished session's repeatable workflow into a reusable SKILL.md

Any Model for VS Code Copilot

Smart Compact — recoverable context cleanup, verified compaction, session navigation and project memory for Pi Coding Agent.

Dashboard for InferenceX™, Open Source Continuous Inference | InferenceX 仪表板

NullFlux — 本地部署的中文 CFD / 科学计算智能体（OpenFOAM · MFIX · LBM · ParaView · 视觉核验 · 论文复现），by LZF

Autonomous AI Company OS — 14 AI agents loop 24/7, ship real software

llmBench is a high-depth benchmarking tool designed to measure the raw performance of local LLM runtimes (Ollama, llama.cpp) while providing

Interactive AI tutor built as a Claude Code skill — teaches with spaced repetition, personalized project examples, and a visual progress das

Production-oriented multi-agent course advisor with deterministic action governance, offline regression evaluation, SSE contracts and observ

Multi-agent exploratory data analysis system with autonomous insights, visualization, preprocessing, and reporting workflows.

A local, open-source alternative to NotebookLM. Chat with your documents using any LLM.

Personal AI agent skills for Claude Code, Copilot, Codex, Gemini & Cursor — turn a PR or idea into marketing content, technical Slidev prese

The last harness you'll ever need.

Make your agents learn from experience. One protocol for weights, harness and routing.

Community benchmark database for running LLMs on Apple Silicon Macs

One-click knowledge system for documents, internal bots, and AI agents

An AI-powered crypto X (Twitter) reply-guy bot: a live tweet feed feeds a scoring engine, an LLM (Claude or OpenAI) writes the reply in a co

Self-hosted AI marketing platform. Like Google Pomelli, but open source, model-agnostic, and actually ships to your social media.