Repository index
6791 AI, ML and Web3 projects · page 9 of 68
- AGENTOPS: AI AGENT MONITORING
Python SDK for monitoring AI agents, tracking LLM costs, and benchmarking. Support for all popular frameworks.
- Sylinko Smart Assistant
Context-aware AI assistant for the desktop that intelligently answers requests and works easily with various LLMs and MCP tools.
- ATOMIC-AGENTS - BUILD AGENTS ATOMICALLY
Framework for building AI agents from minimal components. Python, flexible architecture, 5.9K stars.
- SHADERS FOR WEBGPU
The shaders library provides WebGPU components for React, Vue, Svelte, Solid and plain JS with Framer integration. It includes a CLI tool for generating and compiling WGSL/GLSL shaders and a typed TypeScript API; the project has 1.4k stars.
- LINKEDIN GROWTH FROM TERMINAL
Claude Code skills for LinkedIn: write lively posts, smart comments, and analyze the feed straight from the terminal.
- Media Generation Studio
Platform for creating images and video from a single text prompt with individual model settings and a unified gallery of completed runs.
- Claude-Powered YouTube Toolkit
Eleven Claude skills automate YouTube channels: generate scripts from 21 hook formulas, lint titles/thumbnails together, create edit decision lists, detect retention drop-offs, cut Shorts, and score virality.
- Eve: Framework for Agents
Eve provides a TypeScript library for creating, testing, and deploying autonomous AI agents with tool and memory support.
- AI Bug Hunter
A set of tools for finding vulnerabilities, powered by AI and available without a subscription.
- Strands Agents SDK
A model-based approach to creating AI agents in just a few lines of code.
- GPT Image 2 Prompt Library
The largest prompt library for GPT Image 2: 2000+ examples with previews, 16 languages.
- CODEPILOT: AI AGENT FOR CODE
Desktop client that unifies any AI providers via MCP and skills, controllable from a smartphone. Built on Electron + Next.js, written in TypeScript.
- SQLCHAT: CHAT-BASED SQL CLIENT
SQLChat — interactive SQL client and editor operating via chat interface. Supports multiple DBMS and allows writing queries in natural language with AI.
- RUN VLM LOCALLY ON MAC
mlx-vlm runs and fine-tunes vision-language models (LLaVA, Florence-2, Idefics) on Apple Silicon via MLX. Everything works locally on your Mac without GPU servers.
- RUOYI-AI AI Agent Framework
The ruoyi-ai framework simplifies creating AI agents: it manages multiple LLMs, provides secure knowledge bases, and offers visual workflow orchestration.
- PHP CLIENT FOR OPENAI
Powerful PHP client for working with the OpenAI API. Everything needed to integrate AI capabilities into PHP projects!
- LLM FROM SCRATCH IN PYTORCH
The reasoning-from-scratch project shows how to build a Reasoning LLM from scratch in PyTorch, step by step. Implemented in Jupyter notebooks: architecture, training with CoT and distillation, plus experiments with GRPO.
- AI-POWERED VIRTUAL EXECUTIVE TEAM
OpenExecutive — a single AI executive backed by 8 specialist agents. Stack: FastAPI + Next.js, Claude. 5.8k stars on GitHub.
- Session Recording and Skill Creation
A desktop app captures actions on the screen and, using GitHub Copilot CLI, turns them into an intent and sequential steps, creating a reusable skill for Microsoft products.
- COST-XRAY: SEE CLAUDE CODE COSTS
Proxy tool shows what Claude Code and Codex send to API and calculates cost per request. Supports prompt caching and detailed token analytics.
- HELICONE LLM OBSERVE
Open source platform for LLM observability: monitoring, evaluation, and experiments in one line of code. Supports TypeScript and already has 5.8k stars.
- PGAI: AI INSIDE POSTGRESQL
A toolkit for building RAG, semantic search, and AI applications directly in PostgreSQL. 5801 ⭐
- DeepChat — AI Personal Assistant
DeepChat — a smart assistant that connects powerful AI with your personal space.
- KLAVIS AI: MCP Agent Platform
Integration of tools for AI agents — a reliable and scalable MCP. 5746 ⭐ Python
- tiny-llm - BUILD MINI LLM M1
Course on deploying LLM on Apple Silicon: build your own tiny vLLM + Qwen in just minutes.
- Auto-Generate Loading Skeletons
Framework for creating smooth loading animations with minimal code. Generates skeletons automatically based on the layout.
- SPRITE GENERATION FROM PROMPTS
Agent Skill for generating 2D sprite sheets, transparent PNG frames, and animated GIFs from text prompts. Supports pixel-art style and integration with Claude Code, Codex, and Grok. 5561 stars on GitHub.
- Prompts as Code
370+ cases, 20+ templates, and skills for basic image generation.
- DOCKER WRAPPER KOKORO-82M TTS WITH OPENAI API
Docker image with FastAPI wrapper over Kokoro-82M speech synthesis model. Provides OpenAI-compatible API, websockets for voice agents, and web UI for audio generation.
- OH-MY-HERMES: Smart Agent for All
Easy-to-install Python agent that delivers professional intelligence to any project. All features are available right after a single installation.
- BINARY DEFENSE AGAINST AI AGENTS
BLA is an adversarial defense for binaries, Electron, JS, Python, and Wasm. It poisons context, exhausts resources, and induces hallucinations in reverse-engineering agents like REA. TypeScript, 574 stars.
- RL INFRASTRUCTURE FOR EMBODIED AND AGENTIC AI
Scalable reinforcement learning infrastructure for robotic and agentic systems. Supports real robots, simulators, and VLA models with distributed rollout.
- MCP Tools for iOS/macOS
Server and CLI based on Model Context Protocol, providing tools for automating tasks in iOS and macOS projects.
- holaOS AI Automation
holaOS turns routine tasks into autonomous AI flows, automating repetitive work.
- ZENML: AI PLATFORM FROM PIPELINE
ZenML is a Python framework for creating AI pipelines and agents, unifying data, models, and orchestration in a single platform.
- MMF: Modular Multimodal Framework
Modular framework from FAIR for vision and language research, combining the flexibility of Python and the power of multimodal models.
- MCP Windows Automation Server
MCP server for controlling a Windows computer via the Model Context Protocol.
- FAST MODEL ROUTER FOR AGENTS
Router directs each prompt to the optimal model in under 50 ms, cutting costs by 40‑70 % with a simple endpoint change. Implemented in Go, it has already earned 4174 stars on GitHub.
- BB - Self-Assembling IDE Agent
Project bb is an IDE-agent that can write and improve its own code using TypeScript and LLM.
- SRT to Whiteboard Animation
Converts SRT subtitles into flowing handwriting animations on warm beige paper backgrounds using mask-based stroke rendering.
- HEY-API - CODE FROM SPECS
hey-api turns API specifications into production-ready TypeScript code, saving development time.
- COMFYUI-COPILOT: AI ASSISTANT FOR NODES
ComfyUI-Copilot is a TypeScript custom node that adds AI support to ComfyUI workflows. It automates image generation, offers prompts, and optimizes node chains based on context.
- TELEGRAM + CHATGPT — READY BOT
Popular Python bot with ChatGPT in Telegram: 5.5K stars on GitHub. Simple integration via OpenAI API.
- OPEN-SOURCE AI AGENT FOR COLD EMAIL
Sellbird finds companies and decision makers, verifies emails, writes personal cold emails, sends them and handles replies. Runs on your Claude or ChatGPT plan. 4612 GitHub stars.
- UltraRAG: RAG pipeline framework
Low-code MCP framework for creating complex and innovative RAG pipelines.
- WIGOLO: Local AI Assistant
Wigolo — local search and data collection for AI agents without API keys or the cloud. Free, public beta, 621⭐.
- Obsidian-Mind AI Memory
Obsidian-mind creates a self-organizing vault in Obsidian, giving AI agents persistent memory. It supports Claude, Codex, and Gemini CLI.
- DAFT — Data Engine for AI
High-performance engine for AI and multimodal tasks that processes images, audio, video, and structured data at any scale. Built on Rust ⭐5533
- Autonomous research platform
PRAXIST is an autonomous system for computer‑executable research. It automates literature search, hypothesis generation, experiment planning, and creation of reproducible reports.
- OPEN-SCIENCE AI WORKSTATION
open-science is an open, model‑agnostic AI platform for scientific discovery built with TypeScript. It accelerates experiments and model integration.
- DEVSPACE AI CODING IN BROWSER
The project turns any chatbot (ChatGPT, Claude) into a full‑fledged AI assistant for writing code directly in the browser.
- AI Agent Lifecycle Platform
Next-generation AI Agent Optimization Platform: Cozeloop tackles AI agent development challenges with full lifecycle management from development and debugging to evaluation and monitoring.
- Standalone Claude Coding Workspace
A standalone macOS app providing an integrated Claude coding workspace with Linux terminal, browser, and code editor.
- Designer skill toolkit
A collection of skills from ConardLi: web design, knowledge search, image generation, and more.
- Bindu Live Microservice
Bindu turns any AI agents into compatible, observable, and composable microservices.
- FrontierAgent framework
FrontierAgent is an open framework for building agents with a native CLI‑TUI, ReAct support, and agent team work modes. It launches with a single command on macOS and Linux, requiring no prior installation and no hard dependency on Docker.
- LM STUDIO CLI: MANAGE MODELS
Command line for quickly launching and managing local LLMs in LM Studio. Simplifies working with models without a GUI.
- MOONCAKE - KIMI LLM PLATFORM, 5400+ STARS
Mooncake is the backend platform for Kimi from Moonshot AI. High-performance open-source LLM service in C++.
- Skillhub - Skill Registry
Self-hosted skill registry for corporate agents: versions, RBAC, audit, deployment in Docker/K8s.
- DESKCOMMCRM — OPEN AI CRM
Open-source self-hosted CRM with AI agents and WhatsApp integration, an alternative to Kommo, Octadesk, and Intercom.
- GISKARD-OSS - LLM AGENT EVALUATION
Library for evaluating and testing LLM agents with open-source code.
- Spec-driven development
Spec-driven development for large codebases
- MimiClaw: AI on Microchip
The project lets you run the OpenClaw framework for neural network inference on a $5 microcontroller without an OS, Node.js, or expensive boards. The entire stack is written in pure C and runs directly on the hardware.
- 5IRE: AI-Assistant of the Future
Cross-platform AI assistant with MCP support, local knowledge base, and compatibility with leading providers. ⭐ 5,247 stars on GitHub.
- FALLOW: CODE INTELLIGENCE ANALYSIS
Fallow analyzes TS/JS code: health, complex hotspots, duplication, architecture boundaries, circular dependencies, design‑system drift, and unused code — all from a single graph. Available as CLI, GitHub Action, LSP, MCP, and VS Code extension. Written in Rust, MIT licensed.
- RAG FOR MONOREPOS
AI tool for understanding and editing multilingual codebases through knowledge graphs. 2274 ⭐
- Stitch Skills Library
A skills library for working with the Stitch MCP server, compatible with the Agent Skills standard. It lets you integrate skills with Antigravity, Gemini CLI, Claude Code, and Cursor.
- Plannotator: Visual Code Annotations
Visually annotate and view code‑agent plans and code diffs, share with your team, and send feedback to agents with one click.
- Autonomous CLI Task Agent
Runs command-line agents autonomously in background, only prompting for human approval on uncertain decisions.
- UNIDBG - ANDROID EMULATION ON JAVA
Unidbg allows emulating Android native libraries and experimentally iOS directly on Java, simplifying testing and debugging of code.
- EDIT-BANANA: STATISTICS NOW EDITABLE
Framework for converting statistical formats into editable ones. 5300 stars on GitHub!
- MCP-UI: FUTURE UI DEVELOPMENT
Create modern interfaces using the MCP protocol and TypeScript SDK. Easily integrate AI agents and get flexible UI solutions.
- PPTAgent: AI CREATES PRESENTATIONS
A Python-based agent framework for automatic generation of PowerPoint presentations with a reflective approach.
- Claude + Obsidian Wiki
Persistent wiki repository with AI based on Claude and Obsidian. Automated research, saving, and an accumulating knowledge base following the Karpathy pattern.
- CLAUDE.md for tokens
A CLAUDE.md file that optimizes Claude responses. Reduces redundancy and saves tokens in complex scenarios.
- YouTube Automation with AI Agents
Fully automatic YouTube channel management with AI. Creates, optimizes and publishes videos 24/7 without programming skills.
- Reclip: Media Downloader
A lightweight web interface for downloading videos from any site. Just paste a link and save the content.
- EASYR1: MULTIMODAL RL TRAINING
EasyR1 is a scalable framework for multimodal reinforcement learning training on veRL, delivering high efficiency and ease of use.
- MONKEYCODE: AI CODE ASSISTANT
MonkeyCode is an online AI coding platform for corporate teams that accelerates development and improves code quality.
- ARGILLA - CREATE DATASETS FASTER
Argilla helps AI engineers and domain experts collaboratively create high-quality datasets for training models.
- List of Top AI Tools
A collection of the best tools for working with artificial intelligence: frameworks, libraries, and services.
- CHAT ON STEROIDS: POWERFUL LLM
The project provides local Model Context Protocol capabilities for ChatGPT, fully integrated with the Chrome extension. It supports the Goal, Compact & Resume modes and enables steadfast multi‑agent workflows.
- Human-in-the-Loop Adapter for OpenOPC
Enables zero-modification human intervention in OpenOPC's agentic DAG runtime. Lets operators pause, inspect, and redirect agent workflows without code changes.
- AGENT SKILLS FOR MOBILE SCREENS
The appllama-skills project provides a set of agent skills that turn top‑app templates into native mobile screens. It works with Claude Code and enables generating UI‑code directly from pattern descriptions.
- AUTONOMOUS AI DESKTOP COMPANION
An autonomous AI assistant living on your desktop. Observes the screen, tracks the active window, and proactively interacts via voice and an animated interface.
- Smart Agents with RL
AReaL — a bridge between reinforcement learning and LLM agents. Simple and flexible.
- DSH Sidebar
OMDSH‑dev/DSH-better-sidebar – an extensible sidebar that lets third‑party extensions register their pages. Built‑in file renderer/editor, terminal, chat box, Git interface, and sub‑agent.
- ASTRON RPA: Ready Agent
A ready-to-use RPA toolkit in Java: quickly create agents for automating business processes. Suitable for individual and corporate teams.
- SPARROW: POWERFUL DATA EXTRACTION
Sparrow extracts structured data, executes instructions, and builds agent workflows based on ML, LLM, and Vision LLM.
- OPENAGENT: AI ASSISTANT OF THE FUTURE
openagent – the next level of personal AI assistant: LLM, RAG, and agent loops for computer, browser, and code work.
- Kaku — AI Coding Terminal
Fast terminal with AI-assistant support out of the box. Ideal for AI development.
- Dossier Resume Builder
Self-hosted web app that generates customized resumes and cover letters from your professional profile.
- REACT EMAIL EDITOR WITH DRAG-AND-DROP
A powerful component for creating email templates in React.js without a single line of code. Drag blocks, customize styles, and send!
- XTUNER: TRAINER FOR LARGE MoE
A Next-Generation Training Engine Built for Ultra-Large MoE Models. ⭐ 5138 stars on GitHub!
- JSON_REPAIR - FIX JSON INSTANTLY
The json_repair library automatically fixes corrupted JSON obtained from LLM, API, logs, and user input, saving developers' time.
- TEXT-EMBEDDINGS-INFERENCE - Instant Embedding Inference
Fast inference for text embedding models in Rust, providing low latency and high throughput.
- LOCAL CODE IN KUBERNETES POD
mirrord runs local code as if it were running inside a pod in Kubernetes: real env vars, DNS, network, and traffic.
- Worktrunk: Git worktree management
CLI tool for managing Git worktree, optimized for parallel work with AI agents.
- MOBILE AUTOMATION VIA MCP
MCP server for automation and scraping on iOS, Android, emulators, simulators, and real devices. 5000+ stars!
- Gateway for corporate AI
The fastest AI gateway: 50x faster than LiteLLM, support for 1000+ models, load balancing, cluster mode, limits and overhead under 100 µs at 5k RPS.