Repository index
6792 AI, ML and Web3 projects · page 38 of 68
- AGENTREGISTRY: AI AGENT REGISTRY
A centralized and verified registry for rapid deployment of AI agents. Accelerates innovation and simplifies development.
- CLAUDE-OF-TANKS: 3D BROWSER BATTLE
World of Tanks style Three.js armored sim on Vite. Simulates plate‑by‑plate armor, ballistics, modules, spotting and physics; 110+ vehicles, 16 destructible fields; playable in browser on desktop/mobile; built by Claude/Codex multi‑agent pipeline.
- AI AGENT FOR PENTESTING
Mitsu (EVA) — AI agent for pentesting, automating attacks via structured scenarios and contextual analysis. Supports multiple LLM backends and runs in CLI. 534 stars on GitHub.
- DOCUMENTATION INTO TESTS AUTOMATIC
Truecourse automatically turns documentation into executable tests. When a test fails, it pinpoints the exact docs section where the product differs from the description, helping quickly spot the mismatches between code and the documentation.
- DROIDLINE: ANDROID AUTOMATION WITHOUT ADB
Automate Android from code: tests on real phones, routine scripts, device farm, or AI agent control. Works via Python, Node.js, CLI, HTTP, and Appium. No ADB, USB cables, or root required.
- AI SHORT FILM PROMPTS
203 stars on GitHub. Methodology and prompts behind the zombie short, called one of the best in recent years.
- ICO PS2 Decompilation for PC
A C-based PC port of the ICO PS2 decompilation, enabling the game to compile and run without any original game data.
- MEMT-1.0: LONG-TERM MEMORY FOR LLM
Official MemT-1.0 repository. Implements long-term memory for LLM agents. Supports long context and DeepSeek integration.
- FUD AI: Calorie Tracker
Free open-source calorie tracker with AI nutrition analysis for iOS and Android.
- WHISPLA AI CHATBOT ON RPI
A compact AI chatbot on Raspberry Pi Zero 2W, written in TypeScript. Easy to deploy and chat with LLM.
- VECTORHUB - DEPRECATED REPOSITORY FOR EMBEDDINGS
The archived VectorHub project showcases early work on vector search. The team is now focused on SIE — a self-hosted engine for embeddings, ranking, and document processing.
- MASTER OF CODE AGENTS
Machinist runs AI coding agents as supervised local workflows. It enables supervised execution of agents like Claude Code, Codex and others, controlling their actions via a simple API.
- DEEP SEEKER ON ANDROID
Launch the DeepSeek Harness AI agent on Android without root or Termux — all in one APK with an extension center. Supports local agent execution and API interaction.
- Local-First AI Creation Workstation
A local-first full-stack AI workstation with chat, comics, drama, writing, and learning modules, supporting vLLM, llama.cpp, and Transformers for multi-engine inference with VRAM scheduling and self-healing.
- PREFIRE — IOS TESTS FROM PREVIEW
Library based on Xcode Preview generates Playbook, snapshots and accessibility tests. Supports SwiftUI and UIKit. 526 stars on GitHub.
- LOCAL AI WITHOUT CLOUD
Hob — open-source desktop client running Gemma and gpt-oss locally on Mac, Windows, and Linux. Works fully offline, no data leaves your machine. Free alternative to Claude and ChatGPT.
- UNIVERSAL SDK FOR MCP SERVERS
Concierge — a universal SDK for building modern MCP servers in Python. 534 stars on GitHub!
- OPEN SCIENTIFIC WORK WITH AI
OpenDraft — free open-source tool: 19 AI agents write an article up to 20,000 words in about 10 minutes with verified references. Export to PDF, Word, LaTeX.
- RUNNING DEEPSEEK-V4.1 552B ON A SINGLE PC
Xolotl is an engine for running large MoE models (DeepSeek-V4.1-Flash 552B, GLM-5.3-Flash, Qwen3.8, LFM3-Omni) on a single PC with RTX 4090. It delivers 33 tok/s on agent tasks with 512K context via expert tiering across VRAM/RAM/NVMe.
- AICOMMIT2 - AI GENERATES COMMITS
Aicommit2 is a fast CLI that uses AI models (Ollama, ChatGPT, Gemini, etc.) to automatically generate meaningful commit messages in Git and Jujutsu.
- LYNAVO-DRIVE - YOUR CLOUD FOR PHOTOS
Open-source replacement for iCloud and Google Photos built with TypeScript. Store, sync, and share photos without surveillance.
- PREMIERE-PRO-MCP - AI VIDEO EDITOR
The MCP server for Premiere Pro combines 280 AI tools, a CEP bridge, and UXP support for smart video editing.
- DarkGPT: AI Assistant
A modified version of the AI assistant with extended interaction capabilities and flexible configuration.
- Session Context for Claude Code
CLI + MCP server and /story skill for tracking tasks, issues, handoffs, and roadmap in the .story/ directory.
- CLAGUE — All Tools in One Window
A Svelte web app that brings all dev tools into one interface. The open-source project has already gained 532 stars on GitHub.
- SPOOL — AI SESSION LIBRARY
Collects sessions from Claude Code, Codex CLI, Gemini CLI — with convenient search via ⌘K. Everything is local and under control.
- AUTO-BROWSER - AI GETS REAL BROWSER
Open-source Python agent: gives AI access to a real browser with human control via MCP.
- REQ_LLM Elixir for LLM
A composable Elixir library for working with LLM, built on Req and Finch. Allows easy integration of models into your code.
- VELORN: AI VIDEO EDITOR
Velorn – an AI-native video editor with real timelines, generative workflows, and local agent control.
- EPISODIC AGENT MEMORY
Records the screen locally, builds structured activity frames, and serves them via MCP — without cloud or LLM in the chain.
- TALK WITH ANY ARXIV PAPER
Project talk2arxiv lets you chat with any arXiv article using ChatGPT's power. Via a RAG approach it quickly finds relevant passages and answers with citations. Source code is in TypeScript, open for contributions.
- VIDEO IN MARKDOWN INSTANTLY
Converts video links from YouTube, Instagram, TikTok, X and other platforms into ready‑to‑use Obsidian Markdown notes. Extracts subtitles or local Whisper transcripts, frames, metadata and timestamps.
- Open-Source RAG Marketing Suite
Open-source enterprise RAG solution with local models for knowledge management, AI content creation, and one-click publishing to 17+ platforms.
- AI PERSONALIZED EXAM SYSTEM
🧠 Adaptive system for generating exam questions based on LLM, knowledge graph, and BKT algorithm.
- COMPOSE PERFORMANCE SKILLS
A library of Jetpack Compose skills aimed at improving UI performance and reducing render time. Suitable for agents.
- OPENCODE SWARM: AGENT SWARM
Plugin for OpenCode implementing hub-and-spoke orchestration of an agent swarm with SME consultation, code generation, and QA review.
- Fable 5.1 Worlds Generator
The project generates game worlds through code using Fable 5.1 and TypeScript, enabling creation of procedural landscapes and stories.
- SPOOL - AI SESSIONS LIBRARY
Collects sessions from Claude Code, Codex CLI, Gemini CLI, and others — with search via ⌘K. 530 ⭐
- CUSTOMIZE CLAUDE CODE WITH OP CODE
Use the OpenCode Go subscription in Claude Code—a pluggable middleware tool in Go for seamless integration.
- CODE-ON-INCUS – ISOLATED AGENTS
Each AI agent gets its own container with root, Docker, and systemd, while active protection automatically neutralizes threats.
- OPENCODE-GOAL-PLUGIN - GOALS FOR AI CODER
The plugin adds a Codex‑style goal mode, a slash command /goal, and persists goals between sessions, helping the AI agent focus on the task.
- VIBING-STEAMPUNK BRIDGE ABAP MCP
A bridge between ABAP/AMDP and Model Context Protocol in Go, simplifying vibrational coding.
- ARCHIFY: Architectural Diagram Generation
Architectural diagram generator with dark/light theme toggle and export to PNG, JPEG, WebP, and SVG.
- UNIFIED MCP SERVER FOR AGENTS
The agent project combines multiple Model Context Protocol servers into a single TypeScript solution, simplifying work with AI agents.
- PERCHO: MINIMALIST GUI FOR PI AGENT
Percho provides a clean, desktop interface for the Pi coding agent, using the same engine as the CLI. It supports multi‑session chat, visual tool approvals, and customizable themes for developers.
- AI Photo & Video Search
Searches photos and video frames using deep AI models across any macOS folder. Results are returned instantly with thumbnail previews.
- ADRAFINIL KEEP MAC AWAKE
Adrafinil automatically prevents Mac from sleeping while AI coding agents are working. Supports Swift, works with Amphetamine‑like commands and integrates with Claude Code. 496 stars on GitHub.
- FIRE RED ASR2S SOTA RECOGNITION
Full-featured industrial ASR system with speech recognition, VAD, LID, and punctuation, supporting Chinese, English, code-switching, and singing.
- Hypothesis Discovery Workbench
Python workbench for hypothesis discovery with evidence graphs, synthetic data, and Elo‑ranked debate.
- CATALOG OF AI AGENTS AND TOOLS
ai-agent-tools is a curated set of AI tools, utilities, and resources for devs & creators. Repo >500 stars and ready‑to‑use examples for quick LLM work.
- VLA MODELS FOR CARS
A collection of Vision-Language-Action models for autonomous driving: an overview of past, present, and future directions.
- LLM Serving in Kubernetes
OME is an operator for running LLMs in Kubernetes with smart GPU scheduling and support for SGLang, vLLM, TensorRT-LLM, and Triton.
- MCP FOR DETECTION ENGINEERING
The project provides an MCP server to automate detection rule creation. Integrates with SIEM and EDR, generates Sigma rules and log queries in seconds. TypeScript, 505 stars on GitHub.
- ModelScope Developer Handbook
A hands‑on guide for developers to select, run, fine‑tune, evaluate, and deploy open‑source models, covering RAG, agents, and AIGC from first run to production.
- PROMETHEUS MCP SERVER FOR AI
An MCP server enabling AI agents and LLMs to query and analyze Prometheus metrics through a standard interface.
- CLAUDEXOR - Local AI Controller
Manage multiple Claude/Codex/Cursor/OpenCode subscriptions with a local scheduler. Automatic rotation, shared context, and cross-model review in TypeScript.
- AUTO ENUM GENERATION IN GO
enumer is a Go utility that automatically generates common methods for your enum types: String(), MarshalJSON, UnmarshalJSON and parsing from a string. Just add the comment //go:generate enumer -type=MyEnum and get ready‑to‑use code without boilerplate.
- TERMISH: AI TERMINAL FOR MOBILE
Termish is an open-source SSH/Mosh terminal with a built-in AI workspace for Android and iOS. It combines reliable remote connections with AI-agent capabilities (Claude Code, Codex) in a single app.
- 🤖 WANMAN
wanman - an open environment for running AI agents that operate autonomously on your machine. You watch as agents coordinate tasks among themselves.
- Lyrie.ai — Autonomous AI Cybersecurity Agent
Lyrie.ai — the world's first autonomous AI agent for cybersecurity by OTT Cybersecurity LLC.
- OPENPAPER: Your Smart Researcher
A unified workspace for managing your scientific library: read, annotate, and analyze articles with an AI assistant.
- OCTOCODE — CODE SEARCH WITH MEANING
Useful codebase search and utility for developers. Written in Rust, already 399 stars on GitHub.
- XYBRID: On-Device AI
xybrid — framework for creating applications with on‑device AI in Rust.
- NUDGEBEE: AI IN CLOUDOPS
Unified CloudOps platform with AI‑SRE, AI‑FinOps, and AI‑K8sOps, plus an agentic automation builder — no fragmentation and no model lock‑in.
- PULLMD: CONVERT CONTENT TO MARKDOWN
Self-hosted service converts URLs and files (pages, documents, images, audio, YouTube) to Markdown. Supports PWA, REST, MCP, Claude Code skill, Reddit‑aware, and refreshable share links.
- HERMES ANDROID: CHAT WITH AI
Hermes Android is a Flutter/Dart client for chatting with Hermes Agent directly from your phone or tablet. Connect via local Wi‑Fi or a secure Tailscale tunnel and continue the chat with the AI assistant anywhere.
- PRODUCTION-READY REACT NATIVE BOILERPLATE
Production-ready React Native boilerplate for iOS, Android and Web with Expo SDK 54, React 19.1, Redux Toolkit, TypeScript and AI‑optimized development via Claude/Cursor. Features modern architecture, static web rendering and automated CI/CD.
- DreamServer: Local AI
DreamServer is an open-source project for running LLMs locally with a chat interface, voice control, agents, workflows, RAG, and image generation, operating without cloud or subscriptions.
- NEURO-SAN AI STUDIO
Neuro-san studio – a playful environment for rapid prototyping and testing of neural networks in Python. Suitable for experiments with LLMs and generative models.
- HTML Anything
A local AI agent generates HTML for you — you just publish.
- MCP SERVER: VIRTUAL NUMBER
The MCP server lets AI agents rent a private number in over 200 countries, read SMS codes, and instantly return them via a single API call.
- Framefields Video Engine
Framefields lets developers define videos in TypeScript and renders them directly on WebGPU, enabling real‑time procedural video generation in the browser.
- Continual Learning Framework
HERMIT enables recursive self-improvement for AI models through harness adapter evolution, supporting continual learning without catastrophic forgetting.
- AWESOME-RAG: RESOURCE LIBRARY
A curated collection of resources on Retrieval-Augmented Generation for LLMs. Perfect for developers working with RAG technologies.
- Tossinvest CLI AI Agent
CLI tool for working with Tos투자증권 via an AI agent: full Open API, exclusive WTS features, and an MCP server with automatic key routing.
- CC0 Pose Reference for AI
Free CC0 mannequin pose library with 250+ poses across 5 angles in line art, multilingual names for AI image generation tools.
- AGENTS FOR SOLANA DEVELOPMENT
Skills for agent-based development on Solana with TypeScript support. Accelerate dApp creation with AI assistants.
- OPEN-SOURCE AIaaS PLATFORM
RESTai — an AIaaS platform with Ollama/vLLM support, built-in image and audio generation, visual block programming, and analytics.
- CLOUD CODE: LSP PLUGINS
Marketplace of plugins for Claude Code with LSP server support, expanding the AI assistant's capabilities.
- FASTEST VIDEO ENGINE FOR AGENTS
Kinetograph is a programmatic video engine in TypeScript for AI agents. Uses GPU and FFmpeg to render in seconds. Supports MCP and Hyperframes concept.
- REVDIFF - VIEWING DIFFS IN TERMINAL
A TUI tool for viewing diffs, files, and documents with the ability to add inline annotations. Written in Go, convenient for code review.
- RESEARCH-UNITS-PIPELINE-SKILLS - SMART PIPELINES FOR RESEARCH
The project proposes treating skills as executable units with declared inputs/outputs, acceptance criteria, and guardrails, providing an evidence‑based approach to research.
- AI Surveys on Steroids
edsl — a framework for creating AI surveys and experiments with thousands of AI agents. Ideal for social sciences and marketing.
- LEARN WITH AI AGENT
A recursive learning system with Socratic dialogues and TDD-style exercises right in your code assistant.
- WARREN - Manage Your Agents
Warren is a platform for managing coding agents that isolates, self‑governs, self‑heals, and improves on your infrastructure.
- CONTINUOUS-EVAL - LLM EVALUATION ON DATA
Platform for automatic evaluation of LLM applications based on real data. Enables measuring quality and improving models without manual effort.
- GPUCANVAS: THOUSANDS OF LIVE DOM NODES ON GPU
Library renders thousands of live DOM elements on an infinite canvas via WebGPU. Editor, shaders, and AI agent work with them directly.
- LOCAL AI UNCENSORED
Desktop app with chat, agent, image and video generation. One .exe, no Docker. 518 ⭐
- gitagent - Universal AI Agent
Framework for creating AI agents that live inside a git repository: rules, memory, and tools are stored as files under version control.
- RSSH: SMART SSH CLIENT FOR ALL.
RSSH — cross-platform SSH client in Rust with an AI agent for command autocompletion and log analysis. Supports still mobile devices and desktops, providing a unified experience. 495 stars on GitHub.
- BLITZ STRIKE PENTEST TOOLKIT
⚡ Blitz Strike — universal pentest toolkit for MCP. Methodology: reconnaissance, attack‑surface mapping, source‑to‑sink analysis, live validation. 57 escalation chains, 130‑tool catalog, intelligence data layer; one server — for all agents.
- SYRTIS: MONITOR AI TOKENS
Tracks token usage and quota of 25+ local AI agents in the macOS menu bar. Native Swift with Liquid Glass and a 3D contribution graph shows usage of Claude Code, Codex, Cursor, OpenCode without sending data to the cloud.
- UNITING AI AGENTS IN THE TERMINAL
Runner unites Claude Code, Codex, GitHub Copilot CLI and pi in one terminal, giving each agent its own TUI while working on a shared task. Written in Rust, the project has 201 stars and focuses on agent‑orchestration and agentic‑coding.
- TRADING DESK FOR AI AGENTS
Hyperliquid trading desk with 7 roles and 17 skills for AI agents. Keyless market research, execution via your API wallet and approved tickets. Supports Grok Bot, Claude Code, Cursor, Codex, Gemini CLI.
- WEBROUTER: AGENT STANDARD FOR THE WEB
WebRouter is an open standard connecting AI agents to the web. Allows LLMs to search, extract, and process internet information via a unified interface. Supports MCP and works with any language models.
- Idea Analysis in 10 Minutes
AI agents turn your startup idea into a full validation and go‑to‑market plan in just 10 minutes. They run on Claude Code, OpenAI Codex, and Cursor.
- APPium MCP to the Max
A powerful TypeScript MCP server for Appium. Advanced mobile testing automation with AI is now available to everyone.
- DOTNET-SKILLS: AI SKILLS IN ONE
Installing and using .NET skills for AI assistants: Codex, Claude, Copilot and Gemini.
- ORCA: DEEPSEEK-NATIVE CODING AGENT
Native DeepSeek agent in Rust for coding automation. Manages files, runs commands, and solves tasks in the terminal without extra abstractions.
- MEMTRACE: MEMORY FOR AI AGENTS
Structural memory for AI coding agents with a bi‑temporal graph, MCP‑native integration and zero LLM calls. Works in Cursor, Claude Code, VS Code and more.