Repository index
6792 AI, ML and Web3 projects · page 29 of 68
- MODS FOR CLAUDE CODE
A set of user‑made modifications for Claude Code that extend its capabilities. Adds custom commands, code snippets and hot reload without restarting the editor. Supports working with the latest Claude 3 API versions.
- LIGHTWEIGHT CLI FOR AI ASSISTANTS
Manage AI-assistant capabilities without dependencies. Support for Claude Code, Cursor, Trae and others.
- OPEN COCKPIT FOR OPENCLAW
The web cockpit openclaw-nerve combines voice dialogues, agent kanban board, and workspace control. It lets you launch sub‑agent sessions, view inline charts, and see usage metrics in real time.
- Agent Multiplexer in Terminal
Terminal agent multiplexer. Manage multiple AI agents from one terminal window with convenient screen splitting.
- POKECLAW AI ON ANDROID
PokeClaw — the first local AI agent for controlling Android without cloud or API keys, built on Gemma 4. Control your phone by voice and gestures directly on the device.
- AI COMMITS FOR INTELLIJ
The plugin generates meaningful commits using AI directly in IntelliJ and Android Studio, saving developers' time.
- SHIP-SAFE SECURITY AGENTIC
The ship-safe CLI scanner looks for CI/CD vulnerabilities, agent permission risks, MCP injections, secrets, and dependencies flagged by DMCA.
- AI Teaching Toolkit for Agents
Transforms AI agents from code generators into interactive teachers that explain concepts, guide learning, and adapt to student needs.
- AI Game Dev Desktop App
Desktop app where an AI agent sets up engines (Unreal, Unity, Godot, browser), builds your game, and play-tests it automatically.
- AI System Design: Learning
The repository contains step-by-step materials for designing AI systems based on LLM, RAG, and AI agents. It is suitable for developers and researchers.
- OPEN AI TEAM
OpenSwarm — orchestrator of autonomous AI‑team based on Claude Code CLI. Control via Discord, sync tasks with Linear and use cognitive memory for context.
- CONNECT ANTIGRAVITY TO CODE
The agy-staff project lets you hire Google's Antigravity CLI as a fast Gemini assistant for Claude Code and OpenAI Codex. Built in JavaScript, it already has 708 stars on GitHub.
- SERVERLESS AI CHAT ON AZURE
Template to deploy serverless RAG chatbot on Azure Functions. Uses LangChain.js and TypeScript. Production-ready in minutes.
- 🐍 MODULAR RAG MCP SERVER
A modular RAG system for improving AI answers via knowledge base search. Built on the MCP Server architecture using Skill for step-by-step execution of specifications.
- VOICETREE: Ideal Environment for Agents
Spatial IDE for orchestrating many agents, where you work directly inside the graph, like in Obsidian.
- ThinkWatch: AI Bastion Host
Enterprise bastion for secure access to AI API and MCP with unified proxy service, RBAC, audit, rate limiting, and cost tracking. Supports OpenAI, Anthropic, Gemini, and custom LLMs.
- HAPPIER — Universal AI Client
Cross-platform client for working with Codex, Claude Code, Kimi and other AI agents, providing full end-to-end encryption.
- Agent Memory Specification
Defines a standardized schema and API for storing and retrieving agent memories across platforms. Enables interoperability and versioned memory management for AI agents.
- MOBILEGYM: ANDROID AGENT TRAINING
MobileGym is a browser‑hosted Android emulator for research on mobile GUI agents. The platform provides verifiable evaluation and highly parallel online reinforcement‑learning training.
- OKF AGENT MEMORY GIT
Git-native memory for AI agents: OKF v0.2, sub-300µs BM25 search in memory, embedded MCP server, progressive disclosure. Reduces token bloat by 80% without external databases, pure Go.
- PIG GO-PORT OF PI AGENT
PiG - a faithful port of the Pi agent to Go, preserving the original TypeScript behavior. It allows generating code, refactoring and autocompletion in your IDE, using tree-sitter for fast indexing.
- iOS Simulator Manager
Management of headless iOS simulators with input injection and screen streaming at 60 FPS for iOS 26.
- TRYPOST: SOCIAL SCHEDULING FOR AI
Open-source social media scheduler for AI agents. Lets you create, queue and publish posts via MCP or REST API, and view everything in a single calendar.
- INVISIBLE PLAYWRIGHT MCP FOR AI
Playwright MCP server invisible to anti-bots and captchas. AI agent runs in stealth Firefox, bypasses protection, and automates browser undetected.
- AGENTFILES - Manage agents
Browser and editor for AI agent files directly from Obsidian, with support for Claude, Cursor, Codex and 13+ tools.
- Director: AI Video Studio
Platform for complete video creation using multimodal workflows covering research, scripts, visual development, shoot planning, media generation, and delivery.
- RAPYD ADMIN — LIGHT LARAVEL PANEL
Rapyd Admin — an open-source admin panel for Laravel, created as a lightweight alternative to Nova, Filament and Backpack. It uses Livewire and Blade, allowing you to quickly build CRUD interfaces without extra code.
- SNO STATION SKILLS: CLICK FOR AGENTS
Sno Station Skills provides 31 skills that turn Claude Code and Codex into a coordinated team. It includes peer review between agents, handoff on quota exceed, a one-page morning brief, and a nightly learning loop from your sessions.
- HALO: FRAMEWORK FOR LLM TRAINING
Halo is an open-source framework from White Circle for training large language and multimodal models. It supports agent-based training, async-RL, distributed training, and fine-tuning via GRPO. Written in Python, it has garnered 204 stars on GitHub.
- GPT-TOKENIZER: RAPIDER TOKENIZER FOR GPT
gpt-tokenizer is the fastest JavaScript BPE tokenizer encoder decoder for OpenAI's GPT models (gpt-5, gpt-o*, gpt-4o, etc.). It is a port of tiktoken with features and written in TypeScript.
- HOOP - Unified Protocol Gateway
Hoop is a universal gateway for MCP, LLM, databases, and containers with latency under 5 ms.
- Live Architecture Visualizer
Turn a single JSON config into animated architecture diagrams that render as looping H.264 MP4 videos or live web pages, plus a Claude‑Code skill file.
- Local Knowledge Graph MCP
The MCP server provides persistent memory for Claude through a local knowledge graph, focused on local development.
- ECA: AI Code Assistant
Editor Code Assistant — AI pair programming, editor‑agnostic. Works with any Clojure code editor.
- INCARNAMIND AI CHAT WITH DOCUMENTS
Free desktop app for macOS, Windows and Linux, allowing you to ask questions about your documents and get answers with exact page links. IncarnaMind verifies each citation, guaranteeing its accuracy.
- mcp-searxng - MCP Server for SearXNG
MCP server connecting SearXNG to the Model Context Protocol for fast web search in AI agents.
- SELF-LEARNING QA AGENT
Open-source agent for web and mobile autotests. Write tests in natural language, agent learns from every run, adapts to UI changes, and catches regressions before release. 898 stars on GitHub.
- IMPLEMENTING LLM ARCHITECTURES FROM SCRATCH
The project contains clean implementations of modern LLM architectures in PyTorch. Includes DeepSeek-R1, DeepSeek-V3, Gemma 3, Kimi K2, and Llama 4. Perfect for studying model internals.
- CREATE OH-MY-TAIYIFORGE PROJECT
AI workflow automation plugin for intelligent code generation using Claude and Codex!
- BIGMOEONEDGE - SUPER MOE ON PHONE
The project enables running 284 B Mixture‑of‑Experts models on a device with 12 GB of RAM, using only CPU and llama.cpp without loss of accuracy.
- PROTECTION AGAINST OVERDEFENSE OF AGENTS
HERO‑Anti‑OverDefense — a paste‑in contract that blocks four over‑defense patterns of agents. Works with Claude Code, Codex, Antigravity, Cursor, Copilot, Windsurf and Gemini CLI. Reduces false triggers and increases reliability of AI agents.
- AI Digital Employee Platform
A platform that encodes expert employee knowledge and workflows into configurable AI digital workers, ready for on‑demand deployment, approval, and monitoring.
- MICROGPT-C: GPT IN PURE C
Minimalist GPT implementation in pure C without external dependencies. Train and run models in a single file.
- OPEN AI-AGENT-CONNECTOR FOR INTEGRATIONS
Open-source, self-hosted alternative to Composio, Nango, Merge.dev and Arcade.dev for AI agent integrations. Enables running actions with GitHub, Slack, Hacker News and arbitrary HTTP requests via MCP or REST.
- PERSONAL AI AGENT TEAM
Open-source GrokBot alternative on DeepSeek Harness. Creates bot team with persistent memory, identities and permissions. Extensible via DSH plugins.
- iOS File Manager & Cleaner
Browse files, manage app data, and clean caches on iOS without jailbreak using the worse_query exploit.
- SELF-HOSTED ORCHESTRATOR FOR AI AGENTS
LoopForge is a self-hosted platform for orchestrating coding agents. Manages CodeBuddy, Codex, and Claude Code via scripts with human approval gates. Supports local workflows for coding assistants.
- GORAVEN: Team Agents
Open-source AI-agent platform that reads files, runs code, and calls APIs to deliver ready results for the team.
- Beautiful HTML Templates for Slides
Library of HTML templates letting AI agents auto‑create beautiful presentations.
- SKILLS FOR CLAUDE CODE
Ready-made skills for Claude Code CLI: full-stack development with React, Tailwind v4, Cloudflare, and AI integrations in one package.
- JADX MCP Server for AI
MCP server enabling the JADX‑AI plugin to integrate large language models for code analysis.
- AGENTIC SOC PLATFORM – POWERFUL OPEN
An automated security center with a flexible architecture and agent support, fully open and written in Python.
- DEV-PROXY: SIMULATE FAILURES AND LOAD
dev-proxy simulates API failures, rate limiting and chaos from the command line. It lets you configure fault‑injection scenarios and observe client reactions. Ideal for testing resilience and debugging microservices.
- BAMBOOAI AI DATA ANALYST
BambooAI — AI data analyst that performs transparent reproducible analysis step by step in Python, verifying the evidence behind its conclusions. Supports many models, including local open-weight LLMs.
- LLM ON PS5: LLAMA.CPP NATIVELY
PS5LM ports llama.cpp to jailbroken PlayStation 5. Runs any GGUF models natively on console, including Qwen 3.8. 668 stars on GitHub, written in C++.
- RAG ON AZURE WITH OPENAI AND C#
RAG app on Azure: Cognitive Search retrieves, OpenAI generates answers. Ready template for chatbots and Q&A in C#.
- MCP Server Octocode
MCP server for semantic code search and real-time context generation using LLM.
- CLAUDE SKILLS FOR FOUNDERS
Claude‑skills‑founder — a set of Claude Code skills that help startup founders create product briefs, conduct competitor analysis, shape pricing strategy, and prepare pitch decks and GTM plans straight from the terminal.
- AI Expert Orchestrator
One phrase — and 211+ AI specialists automatically collaborate, delivering a full plan in minutes. YAML, no code, 9 LLM providers.
- AGENTIC FLOW: SWITCH MODELS
Switch between cheap AI models in Claude Code without extra setup. Deploy agents to production.
- AI Agents Learning Path
A structured learning path for AI agents with practice and reading. Trilingual (Traditional Chinese / Simplified Chinese / English) and open for community contributions.
- LLMCORD: Discord as LLM Interface
Connect any model via OpenAI-compatible API to Discord and chat with AI right in chat.
- GOLF: ADVANCED FRAMEWORK FOR AI
Production-ready MCP server framework for building, deploying and scaling the AI-agent infrastructure. Includes auth, observability, debugger, telemetry and runtime. Enables running real-world MCPs that power AI agents.
- LLaVA++: NEW LEVEL OF MULTIMODALITY
LLaVA++ adds support for Phi‑3 and LLaMA‑3 models to the base LLaVA, enabling multimodal chat and image generation. Users gain better text‑and‑image understanding thanks to integration of cutting‑edge LLMs.
- VIBE CODING AUTOMATION
The framework automates code development using AI and computer vision. It supports MCP, game mod creation, and fast repository indexing.
- LLM-Engine: Production Scaling
Open-source engine for scaling and deploying large language models with Python support and flexible configuration.
- ML from First Principles
Educational resource teaching machine learning concepts from first principles, covering inference, CUDA, deep learning, and reinforcement learning with Python implementations.
- PI-POCKET: AGENTS ON MOBILE
Pi‑Pocket is a mobile‑first web app for multiplayer Pi agents, built on the Pi Durable framework. It provides reliable offline operation and synchronises data across devices.
- HERMES-HUD: MIND ON SCREEN
A TUI dashboard to monitor Hermes consciousness directly in the terminal. Watch the AI agent work in real time!
- YOKAI — MODULAR GO FRAMEWORK
Simple, modular, and observable backend framework for Go. 834 stars on GitHub.
- Universal Search for AI
A unified search skill with real‑time support for integrating into AI agents. Simple connection, flexible configuration.
- FROM VIBE TO AGENT ENGINEERING
The project demonstrates best practices for using Codex CLI to transition from vibe coding to agent engineering.
- VIBEUE: Vibe Coding in Unreal Engine
A tool for rapid creation of scripts and logic in Unreal Engine with an AI assistant. Speeds up gameplay development without leaving the editor.
- ANDROID REMOTE MCP - FILE MANAGEMENT
MCP server for Android, optimized for low token consumption, with file download support and automatic setup of Cloudflare and ngrok tunnels.
- Open AI Workstation
Open‑WebUI Tools — a set of 15+ tools and filters that turn Open WebUI into a powerful AI workstation.
- OMEGA‑AI JAVA FRAMEWORK FOR DEEP LEARNING
Omega-AI - Java framework for deep learning with automatic differentiation, multithreading, and GPU acceleration via CUDA and cuDNN. Enables fast building and training of YOLO, diffusion and LLM networks for inference and training.
- INSTAGRAM SKILLS FOR CLAUDE
The project provides skills for Instagram content: captions, carousels, hooks, hashtags and a weekly plan. Materials can be drafted, humanized and published via Publora. MIT license, Creative Content Crafts.
- SVG ANIMATION SKILL FOR CLAUDE CODE
A Claude Code skill that turns any SVG into smooth 60fps animations for React Native and SwiftUI. Solves the "slideshow" problem in agent code: teaches breathing, blinking, gaze following, and living trajectories.
- FLEXIBLE SKILLS FOR AGENTS
Modular skills turn AI agents into full-stack, logic, and multimedia experts, enabling them to solve complex tasks step by step.
- MCP CLIENT FOR LOCAL LLM
TUI client for Ollama with support for all MCP primitives, agent mode, multi-server capability, and streaming output. 749 ⭐
- PG Vectorize: Fast Search
pg_vectorize — a Postgres extension that enables full-text and semantic search in Rust. Easy to integrate, powerful search.
- ADORABLE — AI AGENT FOR CODE
Adorable — open source Lovable, generating code via AI agents. Supports TypeScript and integration with popular LLMs. Enables rapid creation of agents for code generation.
- BANKING API TEMPLATE FOR DEVELOPMENT
The bank-api project is a reference example of a modern, standards-compliant API in C#. It helps quickly start development of a compliant API.
- STYLEKIT: AI STYLE LIBRARY
Open-source visual style library for AI‑generated UI: 148 ready-made sets with design tokens, prompts and MCP support. Enables instant style connection via npm CLI or Agent Skill in Claude/Cursor.
- Autocli for AI agents
Enables AI agents to extract real-time data from 55+ platforms via your Chrome session — no API keys required. Works with ClaudeCode, OpenClaw, and other agents.
- WEBRPC - SCHEMA-DRIVEN BACKEND
webrpc provides a schema‑oriented way to quickly create type‑safe backend services for web applications.
- SLOPO: CODE DUPLICATE DETECTION
Slopo is a code-duplication detector based on embeddings, written in Python. It has garnered 222 stars for accurately detecting repeats.
- TONL - TOKEN OPTIMIZATION IN TYPESCRIPT
TONL is a markup language that optimizes tokens for more efficient LLM operation and reduces query costs.
- MOSS-VL: 11B MODEL FOR VIDEO
The open-weight MOSS-VL series of 11-billion-parameter models is designed for understanding long-form and streaming video in real time. It supports multimodal input and efficient stream processing.
- ONE GATEWAY FOR ALL LLMS
respan-gateway — a single endpoint for all LLM providers. Supports any SDK, any model, and tools. One Rust binary replaces dozens of integrations.
- COMPLETE MIDJOURNEY API IN C#
Unified HTTP interface for Discord, YouChuan and official Midjourney. Processes 1M images per day. Supports face-swap, Gemini and MCP.
- X/Tweet Fetcher — Twitter Parser
Get tweets, replies, feeds, and X/Twitter articles without authentication or API keys. Ready-to-use tool for AI agents.
- IDEA IMPLEMENTATION MCP
The service scans GitHub, HN, npm, PyPI and Product Hunt to verify the reality of AI‑agent ideas before coding. The MCP server has already gathered 754★.
- FRAME VIDEO EDITOR FOR AI
cartcut — video editor for AI agents, built on Electron and TypeScript. Uses ffmpeg for processing, Lit for UI, and supports a macOS app. Demonstrates that open source can outperform commercial tools.
- HELIX: Private AI Agent Fleet
Flagship fleet of AI agents, each gets its own GPU-accelerated desktop. Claude, Codex, Gemini on a private stack.
- SMART NOTIFICATIONS FOR CLIENTS
Cross‑platform tool agent‑notifications delivers smart notifications for Claude Code and Codex CLI: pop‑up alerts, sound signals, click‑focus and webhooks. Works on macOS, Linux and Windows.
- Capso — Screen Capture
Native screen capture and recording utility built with Swift 6.0 and SwiftUI. Free alternative to CleanShot X.
- AGENTSYS AI WRITES CODE
An autonomous AI agent writes code, integrating 26 plugins, 50 agents, and 47 skills for Claude Code, OpenCode, and others.
- PROTECTION OF AI AGENTS FROM THREATS
safelabs-eval — framework for red‑teaming and evaluation of AI agent protection based on the ASI01–ASI10 taxonomy inspired by OWASP. Written in Python, it supports CrewAI and LangChain, enables conducting jailbreak‑tests and vulnerability assessment.
- Fast code search for agents
Precise and fast search of the codebase for AI agents. Uses ~98% fewer tokens compared to grep+read.