Repository index
6791 AI, ML and Web3 projects · page 66 of 68
- AI Agent Rules Sync
Stores a single source of rules for AI coding agents and automatically converts them to AGENTS.md, CLAUDE.md, .cursorrules, and other formats, or creates a new AGENTS.md from scratch.
- Feishu Whiteboard Palettes
Set of 24 ready colour schemes for quickly creating stylish, editable boards in Feishu/Lark. A simple installation script makes them available as an agent skill.
- DiskGenius Professional Edition
DiskGenius Professional Edition provides a full set of tools for data recovery, partition management, and Windows backups. Everything is unlocked with access to advanced disk utilities.
- ComfyUI LoRA Training Suite
Train image and video LoRAs directly in ComfyUI with live previews on NVIDIA and AMD GPUs.
- AlignDev: Code Standards
AlignDev helps frontend teams working with AI assistants create a unified set of coding rules and a SKILL.md file so that Claude Code, Cursor, Copilot, and other agents write code consistently.
- Codex Helper
A lightweight Windows tray proxy that lets you switch Codex CLI between DeepSeek, Tongyi Qianwen, and other Chinese large language models with a single click.
- SVG Animation in Flutter
The Treamz/svg_animate library plays SVG files that contain their own animation (SMIL or CSS keyframes), using the custom vector_graphics renderer from the flutter_svg package.
- Multi-Agent Deep Research Assistant
LangGraph-based research agent with interruptible recovery, dependency-aware parallel execution, citation governance, and 45% cost reduction.
- Secure LLM for Code
An LLM-based system that analyzes AI-generated code for vulnerabilities and suggests automatic fixes. Integrated into CI/CD for continuous application protection.
- Role-Based Plugin Templates
A set of ready‑made templates for creating Codex plugins that adapt to different user roles, simplifying the integration of AI features into applications.
- ARD Resource Search Platform
Neuronto Agentic Resource Discovery indexes public ARD registries and provides federated search, combined with a verified tool index from MCP servers, using hybrid lexical-semantic search and the ARD-Bench test set.
- Inline CSS‑styled CSV
A single plain‑text file that contains CSV data and an embedded CSS stylesheet; opening it in any browser renders a styled HTML table instantly.
- Local AI Media Studio
Windows desktop app for offline image/video generation, editing, and LoRA management using local models.
- Rust Screen Capture Tool
Windows-only screen capture, recording, and annotation tool written in pure Rust, targeting Snagit feature parity.
- Adding subtitles online
The service lets you upload a video and automatically overlay subtitles directly in the browser, with no extra software needed and completely free.
- Hermes Agent for PC
Local application for launching AI agents based on the Hermes model via Ollama. Provides a convenient interface for managing models, workspace, and executing tasks.
- Easy Stable Diffusion Tuning
njgymb/diffusers-sculptor offers a fast, flexible framework for fine‑tuning Stable Diffusion through simple configuration files, lowering resource demands.
- MedSkillOS — Medical AI Agents
A library of ready-to-use AI agent skills for clinical and biomedical tasks: extracting data from EHRs, analyzing medical images, and supporting decision‑making.
- MusicGen Pro Trainer
This 2026 HTML guide shows how to quickly fine-tune the MusicGen model to your own musical styles using ready-made notebooks and a web interface.
- NLP Trainer for LLM 2026
Platform for fine-tuning large language models with automatic hyperparameter selection and visual metric tracking.
- Mathematical Riddle Generator
The project uses a fine-tuned GPT-2 to create unique mathematical puzzles and provides them as an interactive HTML page.
- Automatic LoRA Docker Pipeline
The project automates building Docker images for training LoRA models in Alibaba Cloud, providing a CI/CD pipeline, GPU support, and publishing to a container registry.
- Voice layer for agents
Self-hosted real-time voice interface compatible with any LLM models, designed for integration into PWAs and terminal agents. Provides low-latency speech recognition and voice synthesis via WebSocket API.
- AI Clipboard Manager for macOS
Open-source clipboard manager with history, OCR search, AI translation, and a bring-your-own-model agent. Native Swift, zero dependencies.
- Lossless Scaling for Steam Deck
The project provides tools for lossless frame scaling in games, using AMD FSR, NVIDIA NLS, and its own LSFG 3.0/3.1 algorithm to increase FPS and reduce latency on Windows 11, Linux, and Steam Deck.
- Xiaohei 2.0 Codex Skill
The project provides the Xiaohei 2.0 Codex skill for generating illustrations of real objects and long scroll stories in Chinese, simplifying visual content creation.
- AI Agent Recovery Controller
Open-source tool for managing execution and recovery of AI agents that use code and external tools. Ensures reliable task execution and automatic rollback on errors.
- AI Pain Pattern Experiments
Provides code, data, and results from testing open AI models with a pain pattern enabled.
- macOS‑style C++ UI Framework
A C++ framework that replicates macOS native UI components for desktop applications.
- Rust Retro Launcher Framework
Rust framework for building cross-platform retro game launchers with emulator integration and game library management.
- Chat templates for LLM
Flexible and programmable templates for preparing training and inference data for language models.
- Minimal LiteLLM Gateway on Rust
A lightweight Rust gateway compatible with LiteLLM, designed for integration with coding agents. It provides fast access to LLMs via a standard API.
- LLM course from scratch
Practical course on 23 Jupyter notebooks where you build tokenizers, attention, MoE, RLHF, inference, evaluation, and distillation of modern LLMs in PyTorch.
- AI Workspace with SuperCompress
Open-source TypeScript workspace enabling humans and AI agents to collaborate using SuperCompress context compression.
- Speculative Decoding Training Framework
ASpec accelerates LLM inference by training efficient draft models for speculative decoding with high-throughput distributed training.
- Knowledge for AI Systems
Extracts and parses structured data for AI agents and RAG systems.
- Image AI Generator 2026
Powerful desktop application for creating high-quality images, photorealistic renders, and creative concepts using AI.
- 🐍 MEMPALACE - BEST AI MEMORY
Free memory system for AI that wins benchmarks.
- MCP RAG Server for OCPP
The project implements an MCP server using Retrieval-Augmented Generation to work with OCPP 1.6 and OCPP 2.0.1 standards, providing fast access to EV charging specifications and recommendations.
- Code Unnati 3rd Year
The repository contains study materials, lab works, and projects of third-year AI students at Pallavi College of Engineering, created under the Code Unnati program.
- ComfyUI H3 Video Upscaler
Upscales MiniMax H3 clips via latent upscaling and tiled guided re-draw with x2 LoRA.
- Self-Hosted RAG Search API
A self-hosted API that performs web search, extracts content, generates embeddings, and produces answers via retrieval-augmented generation.
- Neural network for Anima LoRA
A universal tool for creating Anima LoRA models in three stages: from extraction to training.
- Stable Diffusion Pro
Stable Diffusion Pro is a local image generator based on SDXL with ControlNet and LoRA support. Premium subscription is activated for free for a lifetime lasting through 2026, with no trial period limits.
- KeyType: AI Autocomplete
KeyType is an open-source Swift project that adds system-wide AI autocompletion to macOS. It works like Cotypist, predicting the next text in any application.
- Vigils — AI Agent Control
Vigils is a local control panel for AI agents that lets you monitor their actions, approve important operations, and keep secrets safe. It is built with Rust, Tauri, and a Chrome MV3 extension.
- RAG System for Java
A RAG system combines loading multi-format documents, semantic search, and Q&A, focused on Java interview prep and LeetCode Hot100. It uses an LLM when knowledge is lacking, storing verified answers in the knowledge base.
- Campus RAG Knowledge Base Showcase
Interview demo of a campus RAG system with architecture docs, tech specs, and Docker Compose deployment configs. Business source code excluded.
- SecSkills PTES Module
Professional pentest skills module for Claude Code, aligned with the PTES standard. It covers reconnaissance, exploitation, post-exploitation, and antivirus evasion, with regular updates for improved effectiveness.
- INVISIBLE AGENTS FOR YOUR PC
invisible_dots — open-source,self-hosted alternative to OpenAI Dots. Each agent runs in an isolated VM on PC with a full desktop, shell, files, and memory, preserving skills and staying undetectable to antibots. Any models via OpenRouter.
- ComfyUI Nodes for Open-H3-Regenerate
Provides ComfyUI nodes, workflows, and a converted LoRA for the Open-H3-Regenerate-2K-IC model.
- Anima — Hardware OS
Anima — an open agentic OS that makes any hardware intelligent: a unified API for sensors, actuators, and edge-AI.
- Ore-Code: AI Workstation
Ore-Code is a desktop development environment based on the DeepSeek model, offering intelligent assistance with writing, debugging, and refactoring code.
- OmniForge AI Toolkit
A Go-based collection of large language model utilities, Retrieval-Augmented Generation pipelines, autonomous agents, and multilingual service implementations.
- AI-Native Application Framework
CRISSP provides a structured framework for building applications with AI capabilities embedded natively from the ground up.
- TripoSplat: 2D to 3D Gaussians
TripoSplat converts a single image into a variable number of high-quality 3D Gaussians. It was developed by the TripoAI team in Python.
- Disk Drill Pro: Data Recovery
Professional tool for recovering deleted files, photos, and videos from any drive in Windows, with support for deep scanning.
- One-Click Samsung FRP Bypass
The tool allows bypassing FRP protection on Samsung devices with one click, preserving user data. It also supports changing CSC without data loss.
- Airplay Plugin for Omarchy
The plugin adds Airplay media streaming support to the Omarchy application, enabling audio and video transmission to compatible devices.
- Local Windows AI Agent
A Windows AI agent that runs locally from one .exe, with no command line or complex setup required.
- 3D Real Estate Scene Builder
Converts real-estate floor plans and photos into interactive, walkable 3D scenes using Python.
- HiDream O1 nodes for ComfyUI
HiDream O1 nodes for image generation and LoRA training in ComfyUI pipelines on Python.
- Accelerated Text Generation
Fast and lossless output of language models using double diffusion decoding.
- Claude Skill Set for Self-Study
A collection of ready-made Claude skills that help learners develop deep understanding and learn to think critically.
- Premium Data Recovery
Stellar Data Recovery Premium is reliable software for recovering lost files from Windows drives and external media. It includes tools for recovering and repairing photo and video files.
- Minimal RAG Framework
A simple and lightweight framework for building Retrieval-Augmented Generation applications in Python. Allows quickly indexing documents and obtaining answers with context.
- LoRA Repairs Splat Far Views
A LoRA adapter for Qwen-Image-2.1 that fixes distant views in single-photo 3D Gaussian splat reconstructions.
- Edge-Optimized LLMs
The project provides tiny language models specially compressed and accelerated for running on resource‑constrained edge devices. It supports inference on CPUs and mobile chips with minimal latency.
- Improving Research Questions
A portable PowerShell agent skill that analyzes the original question and suggests more precise variants, improving the quality of research search.
- Echo-Infinity: AI Video
Official repository of the Echo-Infinity paper, presenting a trainable evolving memory for generating infinite video in real time.
- AI Code Ontology
An ontological wiki‑base of code with a knowledge graph, semantic search, and a RAG chat, supporting an agent aware of code‑change impact. A cross‑platform Electron application.
- Training GPT for $50
The project shows how to train your own GPT-like model on a $50 budget using accessible cloud resources and open-source code.
- Interactive 3D AI Agent World
Watch AI agents live, work, and teach in a self‑hosted 3D simulation powered by local models.
- WebHawk Security Scanner
Performs authorized enumeration and exploitation of web servers using Python-based modules.
- Local AI Meeting Assistant
Local-first AI meeting notes for Mac with on-device transcription, speaker diarization, and auto-generated minutes linked to audio quotes. Supports Tencent Meeting, Feishu, and Zoom without uploading recordings.
- Autonomous drone against flies
Flyguard uses a drone equipped with a laser system and computer vision to detect and neutralize horned flies on farms, reducing livestock stress and productivity losses.
- Claude Code SDLC Skills
Markdown skill files that guide Claude Code through planning, implementation, testing, and review during the SDLC.
- AI-powered Docs Search
Embed an AI chatbot and site search into any documentation site with a single <script> tag, providing streaming answers with source citations.
- Download manager for browsers
Extension for Chrome, Edge, and Firefox that speeds up downloads via dynamic file segmentation, supports resuming interrupted downloads, and manages a download queue.
- Personal RAG Knowledge Assistant
Python personal RAG assistant for custom knowledge bases, traceable citations, and evaluating AI-generated dev doc retrieval; a DocsGPT preview.
- Offline LLM Emergency Assistant
Runs a local LLM with 200 GB offline knowledge library on macOS, providing cited answers without internet connectivity.
- RAG Study Helper
The project provides a full set of tools for studying and applying Retrieval-Augmented Generation methods in Java, including document indexing, answer generation with citations, and quality evaluation.
- Media from AI in Minutes
A skill set for generating images, video, and audio via AI agents. Works with Claude Code, Cursor, and Gemini CLI.
- ChatGPT LaTeX Sidebar Editor
Open a LaTeX paper in the ChatGPT sidebar, add annotations, and review changes side‑by‑side with your conversation.
- Visual Review for AI Agents
Renders agent plans and decisions as an interactive local web page, capturing human feedback for Claude Code and Codex workflows.
- Manga Panel Layout Selector
ComfyUI custom node that lets you select from 196 manga panel layouts and populate them with characters, scenes, and dialogue for AI manga generation.
- Sponsorship analysis platform
The platform collects real‑time World Cup match data, analyzes textual signals from open sources, forecasts sponsorship ROI with uncertainty quantification, and suggests action scenarios.
- Poseidon M5Stack Firmware
The Poseidon firmware turns the M5Stack Cardputer-Adv into a multifunctional device with Wi‑Fi, BLE, sub‑GHz, LoRa, BadUSB and various attack capabilities, and includes six ready‑made interface themes.
- LabLLM on macOS
LabLLM is a native macOS app where you can build the architecture of a tiny language model, train its weights, and see the model emerge from scratch on Apple Silicon using MLX.
- SmartOpenMAIC Interactive Class
An open platform for rapid deployment of a multi‑agent interactive classroom, where each agent acts as a virtual teacher or student, delivering an immersive learning experience with a single click.
- Llama2 Assembly
A minimalist inference engine for Llama2 implemented in pure assembly for maximum performance and low resource consumption.
- Personal AI agent with memory
Self-hosted AI agent with vector memory, Composio tools, and Telegram integration.
- vLLM for RTX 2080Ti
Optimized vLLM runtime for two RTX 2080 Ti 22 GB GPUs with NVLink, enabling local inference of 27B and 31B parameter models with >100 tokens/sec throughput and up to 262K token context.
- Self-hosted LLM gateway
A single Go binary turns your Mac and Linux machines into a private inference cluster featuring multi‑machine routing, sharding via llama.cpp‑RPC, and support for the OpenAI and Anthropic APIs.
- Inference Agent Tokenmaxxing
Agentic-in/inferoa is a framework for creating agents optimized for inference with Tokenmaxxing technology, providing high efficiency when working with large language models.
- Dual-Clock Euler Sampler
Dual-clock Euler sampler for generating audio‑video content using MiniMax‑H3 Turbo LoRA in ComfyUI.
- Persistent Agent Memory Store
Stores long‑term memory for AI coding agents as a hierarchical file tree, enabling retrieval and persistence across sessions.
- Unified Memory for AI Tools
Store data once — use it in Claude, ChatGPT, Cursor, and any MCP client via self‑hosting on Cloudflare’s free tier.
- Anima LoRA Factory
A graphical interface for simplified LoRA adapter training using modern Anima diffusion models. It lets you quickly configure styles and characters without deep coding knowledge.
- Portable VoxCPM2
Portable VoxCPM2 build for Windows, providing multilingual speech synthesis in 30 languages, voice cloning, and one-click LoRA adapter training.