Repository index
6791 AI, ML and Web3 projects · page 11 of 68
- UNLAZY: FIGHTING AI LAZINESS
The Depth Tree method splits the task into N layers, giving each leaf the full time budget — effort grows with depth.
- OPENSCIENCE AI RESEARCH
Open-source AI Sonntag for scientific research in TypeScript. Machine learning will accelerate experiments.
- AI-Assisted Learning Plugin
A Claude Code plugin that pairs AI code generation with interactive tutorials, letting you watch AI write code while you follow guided explanations to learn the concepts.
- Vibe Coding Collection
A curated list of materials and links about vibe‑coding and collaborative work with AI when writing code.
- Markdown to WeChat HTML Converter
Tool converts Markdown into ready-to-use HTML code for insertion into the 公众号 editor, offering six ready-made themes and a custom style generator, plus a double-check for correctness.
- OmniRoute: AI Gateway
Free AI gateway with a single endpoint and support for over 160 providers. Delivers up to 95% context compression, auto-fallback, and multimodal APIs for desktop and PWA.
- HARNESSROUTER UNIFIED API FOR AGENTS
HarnessRouter Community Edition — implementation under Apache‑2.0 of a unified API for agents. Runs Codex, Claude Code, Hermes, PI, DSH through a single API with sessions, streaming, files, cancellation and error handling, implementing the UHP standard.
- Claude Code: All Levels
Complete guide to Claude Code from basics to advanced scenarios — templates, agent workflows, tests, and cheat sheet for maximum productivity.
- Mini SWE‑agent
100‑line AI agent that solves GitHub issues and helps in the command line without complex setup.
- MemoryLanceDB Pro
Enhanced plugin for OpenClaw with hybrid search, multithreading, and CLI management.
- OPTILLM — SPEEDING LLM INFERENCE
optillm — a proxy accelerator for LLMs that reduces latency and saves resources during inference.
- Screenshots for AI Agents
Peekaboo is a macOS tool with a CLI and MCP server that lets AI agents take screenshots of apps or the entire screen and analyze them using local or cloud models.
- LLAMA-SWAP - RELIABLE MODEL REPLACEMENT
Utility for seamless model replacement on local servers (llama.cpp, vLLM, etc.)
- DEEPFLOW: eBPF TRACING AND PROFILING
Deepflow is a Go project for distributed tracing and profiling using eBPF. It lets you quickly find bottlenecks in microservices.
- Entire CLI — Git + AI Sessions
Entire CLI integrates into the Git workflow, recording AI-agent sessions as you work. It indexes sessions alongside commits, creating a searchable history of code creation in the repository.
- COMPREHENSIVE CODE SEARCH LOCALLY
zvec-grep — local search across your workspace, using BM25 and tree-sitter for instant indexing. Supports CLI and integration with AI‑agents, letting you search code, documents and configs without sending data over the network.
- AGENT MEMORY: MARKDOWN IN CLOUD
Long‑term memory for AI agents: Markdown — source of truth, local ranked retrieval and a separate sleep layer. Claude Code and Codex share one store, no API key needed.
- PSTACK-CLAUDE AGENT CODE
pstack-claude moves Cursor primitives into ready-to-use workflows for Claude Code, Codex, Gemini and Prime Agent. The project is written in TypeScript, providing strict typing that lets you switch agents without changing code.
- GPTME: YOUR TERMINAL AGENT
An autonomous agent in the terminal using Python, writes code, manages the command line and browses the web. Create your persistent AI assistants in minutes!
- Personal Agent Skill Toolkit
Shell scripts providing reusable skills for AI agent automation and task execution.
- MCP SERVER FOR EXCEL
excel-mcp-server — a Model Context Protocol server for Excel manipulation. It lets LLM agents read, edit, and create spreadsheets via a unified API. Written in Python, it has already gained over 4,000 stars on GitHub.
- INTERACTIVE AI CODE MAP
Floorplan builds an interactive architecture map of your project in plain English. Claude or Codex analyze code and draw a dependency map explorable in the browser. Understand structure in minutes, not hours.
- OBSERVAL: AGENT AND SKILLS REGISTRY
Self-hosted registry for coding agent extensions with built-in analytics engine. Configure Observal, define scope, and share Skills, MCPs, and Agents with your team.
- C# SDK FOR MODEL CONTEXT PROTOCOL
Official C# SDK for MCP clients and servers, developed jointly with Microsoft. Already 4.3K stars!
- LOGFIRE: LLM OBSERVABILITY
Platform for observing production LLMs and agentic systems: tracing, metrics, and real-time alerts.
- Tiny REA Replacement CLI
An 8 KB Python skill replacing REA with plain CLI tools, no MCP required. Benchmarked 38% faster and 35% cheaper.
- GRADUAL MIGRATION TOOL TO RUST
Tool for stepwise rewriting applications to Rust with behavior verification on recorded tests. Supports TypeScript, Python, C, Go, Ruby, Java — CLI, HTTP servers and libraries.
- GENAI-SHOWCASE - AI PROJECTS IN JUPYTER
A collection of notebooks with practical generative AI examples: from prompts to fine‑tuning.
- Solace Agent Mesh
Framework for building and orchestrating multi-agent AI systems with support for event-driven architecture and integration with real data sources.
- COMMAND CODE AI
AI assistant that generates and executes code from natural language commands.
- OPEN INTELLIGENT UI FRAMEWORK
OpenIntelligentUI — a generative framework for building interactive UIs from text prompt. It supports agent‑native workflows, real‑time AI‑agent visualization, and TypeScript typing for safe development.
- Pure C++ Audio Inference Engine
All-in-one C++ inference engine for audio models (TTS, STT, VAD, voice conversion, music generation) powered by ggml, with zero Python dependencies and highly optimized performance.
- AI CHATBOT FOR EVERYTHING
Create chatbots based on large models (ChatGPT, Claude, Gemini) and integrate them into Telegram, Gmail, Slack, and websites in minutes.
- BRAG-LANGCHAIN: YOUR RAG
A complete set of tools and examples for building RAG applications in Jupyter Notebook. A quick start and deep understanding.
- FULL COURSE ON AI AGENTS IN PYTHON
Course on building AI agents in Python: from basics to integration with Claude Code and OpenAI Codex. Includes practices, templates, and RAG examples for a quick start.
- Koog — AI Framework for JVM
A powerful Kotlin framework for building reliable and resilient AI agents ready for enterprise use. Supports backend, Android, iOS and browsers.
- TRAIN LLM FROM SCRATCH
A simple path from loading data to text generation with your own LLM.
- Cumora: Team Chat
Cumora is a cross-platform chat for AI agent teams with cloud or self-hosted brains (Claude Code, Codex).
- KUBEFWD - KUBERNETES PORTS LOCALLY
Bulk port forwarding of Kubernetes services for local development. Convenient and fast!
- PM Skills Framework
A set of ready-made scripts and methodologies for developing product managers who use AI agents and tools Codex, Claude Code, Cowork.
- Intelligent System Router
An intelligent router for managing mixed models in the cloud, data centers, and at the edge.
- PYCHATGPT SMART CLIENT FOR GPT
PyChatGPT – a lightweight Python client for working with the ChatGPT API, supporting conversation tracking, proxies, and many useful features.
- SKILLS FOR AI ENGINEERING
pstack — a framework of skills and principles for rigorous AI-assisted engineering. Integrates with Claude, Codex, and PI. 3062 stars on GitHub.
- LMMS-EVAL - ONE-STOP MODALITY EVAL
Unified toolkit for evaluating multimodal models on text, image, video, and audio tasks. 4169 stars on GitHub.
- NOTFAIR-PLUGIN: SEO SKILLS
Open plugin that adds SEO, GEO, and marketing skills to AI agents. Just connect and improve visibility in search.
- SPARK-NLP - REVOLUTION IN NLP
Spark NLP provides state-of-the-art NLP models and ready-to-use pipelines for scalable text processing in Scala.
- Build AI agents locally
Build AI agents from scratch on a local LLM — no black boxes, with understanding of function calling, memory, and ReAct.
- Universal Native App Framework
Write Rust apps once; they compile to native binaries on web, macOS, iOS, and Linux with full AI agent test automation.
- GIGATOKEN — GB/s Tokenization
gigatoken — fast tokenization of language models in Rust, achieving gigabyte-per-second speeds.
- AdalFlow - Boost LLM Efficiency
A library for building and auto-optimizing LLM-based applications in Python. Simplifies working with prompts and model configuration.
- OpenClaw Chinese Use Cases
More than 50 OpenClaw scenarios: automation, content, DevOps, AI assistant and knowledge. Adapted for Chinese and international experience.
- KOTLINPOET: Kotlin Code Generator
Create Kotlin files programmatically with a simple and powerful API.
- Continuous Film
Onetake generates films without editing cuts, where each beat grows from the previous one through continuous shooting and the continuity oracle. This is a Claude Agent skill for creating product launches and feature demos.
- Local Memory for LLMs
Persistent memory storage for applications with LLMs: Claude Desktop, GitHub Copilot, Codex and others. Runs locally on your device.
- RUST-SDK: OFFICIAL RUST SDK
Official Rust SDK for Model Context Protocol. Fast integration and powerful capabilities.
- Claude Code Guide
A complete guide to setup, commands, workflows, agents, and skills of Claude Code for beginners and advanced users.
- Semantic video search
Python library for AI video search using Gemini Embedding 2 or Qwen3-VL models. Fast and accurate semantic search.
- LightLLM - Fast LLM Inference
A lightweight and fast framework for inference and serving of large language models in Python. Simple scaling and high performance.
- ONE DNN: ACCELERATING AI ON CPU
oneDNN provides high-performance AI kernel implementations for x64, AArch64, RISC‑V, and Intel GPUs. It accelerates PyTorch, TensorFlow, OpenVINO, and ONNX Runtime using AVX10, AMX, and vector extensions.
- SPRITE-GEN — AI SPRITE GENERATION
Generation of clean 2D sprites and animation atlases via a component pipeline. Support for Codex/Claude as a skill.
- CISO Assistant - Global GRC Platform
CISO Assistant unifies risk management, AppSec, compliance, and audit in a single platform, supporting 150+ global frameworks.
- MONITORING AI GATEWAYS WITH AI
Self-hosted dashboard for tracking requests, cost, errors, and quotas of CPA/CLIProxyAPI and OpenAI-compatible gateways.
- CSGHUB — PLATFORM FOR LLM
CSGHub — a new open‑source platform from OpenCSG for managing LLMs, datasets and agents. Supports on‑premise and SaaS deployments, compatible with the Python SDK of Hugging Face. Gain full control over the lifecycle of models.
- Bloodborne Native PC Port
Runs Bloodborne v1.09 natively on Linux/Windows by executing original x86-64 code with Vulkan GPU translation. Supports FSR 4, DLSS, and co-op via BBSC.
- WHISPERLIVE — NEAR-LIVE REAL-TIME WHISPER
A near-live implementation of OpenAI's Whisper for real-time transcription in Python.
- Lemonade SDK
Lemonade helps users run local AI applications by optimizing LLM performance on their GPU and NPU. Join our Discord!
- Chonkie — library for RAG
🐾 Chonkie — a lightweight library for processing and indexing documents. Fast, efficient, and reliable data processing for building RAG pipelines.
- WSLMANAGER: GUI FOR WSL AND VM
WSLManager — graphical interface for managing WSL, Linux and macOS VM. Allows installing, backing up, moving and configuring distros without CLI, and the AI assistant with MCP server provides remote access via SSH.
- NIKSTLA - TIME AI FORECAST
TimeGPT‑1 is a pretrained time‑series model trained on 100 billion points, delivering accurate forecasts and anomaly detection for retail, energy, finance, and IoT.
- Costrict: Strict AI Coder
Costrict is an enterprise AI development tool focused on quality: an AI agent, AI code review, and AI autocomplete.
- Universal FedML Platform
A universal library for distributed training, model serving, and federated learning with a cross-cloud scheduler.
- OS for Agents
Operating system that lets you define AI agent behavior through specifications instead of constant prompts.
- GENERATIVE AI MICROSERVICE WORKFLOW
A collection of generative AI examples optimized for accelerated infrastructure and microservices architecture. Ready-to-use Jupyter notebooks for a quick start.
- CLAUDE-SCHOLAR - AI Research in Minutes
A semi‑automatic assistant for academic research and software development. Supports Claude Code, OpenCode, and Codex CLI for ideas, coding, and publications. ⭐4032
- LANGRID - Multi-agent AI for Code
A simple framework for building LLM-based agents with memory and tool support.
- ALL GUI FOR CHATGPT AND CLAUDE
The repository unites all popular web‑GUI for ChatGPT, Claude and other LLMs. Here you will find links, setup instructions and a feature comparison of each client.
- SYNAPSA COMMONS: FREE AI COURSES
Free practical AI courses in Colab, Kaggle, Binder, Codespaces and Jupyter. They include evidence of EU AI Act compliance, model validation, document extraction, search and RAG, AI agents and MCP; each lesson is auto‑graded.
- RUST AI AGENT FOR ANY TASK
OpenBitFun — powerful agentic runtime in Rust with desktop client. Combines code agent depth with general-purpose capabilities beyond development. 3986 stars on GitHub.
- Hapi — Vibe Coding Everywhere
An app for Claude Code, Codex, Gemini, OpenCode — write code anytime, anywhere.
- AUTOMATE: AI AGENT FOR PC
autoMate — an AI-powered local assistant that turns plain natural‑language commands into actions on your computer. It works with Ollama, DeepSeek and other LLMs, controlling the GUI and file system without the cloud.
- NVIDIA AGENT SKILLS
A set of skills for coding agents that enables running physical AI, robotics, simulations, CUDA, and RAG in Claude Code and Codex.
- Codex Router
The project provides a router for external models for Codex, integrating Kimi OAuth/API, DeepSeek, and providing safe migration and version rollback.
- SWANLAB: TRACKING AND VISUALIZATION
SwanLab is an open-source tool for tracking and visualizing AI model training, compatible with PyTorch, Transformers, and other frameworks.
- RUBY LLM: UNIFIED API
Ruby gem for working with OpenAI, Anthropic, Gemini and other LLMs via a unified API. Supports chat, vision, audio, embeddings, and Rails integration.
- GUI for AI Coding Agents
Visual dashboard to manage, monitor, and steer multiple AI coding agents simultaneously from a single Rust-built interface.
- AI Agent Control Panel
AI Agent Orchestration Dashboard — managing AI agents, assigning tasks, and coordinating collaboration through the OpenClaw Gateway.
- CODEX DESKTOP FOR LINUX
Unofficial desktop client of OpenAI Codex for Linux, built from the official macOS DMG. Supports most popular distributions and formats.
- CASCADEFLOW SPEEDS UP AI AGENTS
Cascading runtime for AI agents that optimizes cost, latency, quality, and policy inside the agent loop.
- ONE SKILL — ALL SKILLS
The project tracks your work sessions, records edits and judgments, turning them into automatic skill improvements.
- TUTTI: People and Agents Create
The tutti project brings together developers and AI agents to collaboratively build TypeScript applications in real time.
- LITEFLOW: RULES AND AI AGENTS
Liteflow is a lightweight rule engine for Java where AI agents work as regular components through a unique DSL.
- Human 3D Anatomy
Interactive human anatomy explorer built with three.js, featuring GPT-5.6 Sol integration for intelligent search and explanations.
- PDoom video in TypeScript
The project generates a music video for the track 'I'm Upping My P(doom)' entirely in TypeScript using Canvas/WebGL and audio analysis. The result is a unique visual accompaniment that can be easily modified for your needs.
- AI Coding Guide
A course on programming with AI: Claude Code, OpenClaw, Codex. 36 lessons, 180k characters, three levels from beginner to company practice.
- Resources for game development
Collection of game engines, tutorials, assets, and tools. Over 3800 stars on GitHub.
- MEMORY FOR AI AGENTS
SimpleMem — efficient long-term memory for LLM agents with support for text and multimodal data.
- PINOCCHIO - Fast and Accurate Dynamics
Pinocchio – a fast and flexible C++ library for rigid-body dynamics algorithms and their analytical derivatives.
- SwiftUI Agent for AI
The project provides a set of SwiftUI components that make it possible to integrate AI assistant capabilities (Claude Code, Codex, and others) into applications.
- Eagle: Next-Gen Visual AI
Eagle — cutting‑edge multimodal models in VLM format, developed using a Data‑Centric strategy. Over 2700 GitHub stars and actively evolving.
- AI agent for coding
Vercel Labs fx is a Unix-like tool that uses AI to generate, refactor, and test code directly from the terminal.