Repository index
6792 AI, ML and Web3 projects · page 31 of 68
- AFFILIATE-SKILLS 50 AI SKILLS
A set of 50 AI agent skills for affiliate marketing: trend research, post writing, infographics, landing pages, and deployment.
- TypeScript Style Guide - Code
A set of rules and best practices for writing clean and maintainable TypeScript code. Quick start and deep integration.
- ISOLATED ENVIRONMENT FOR AI AGENTS
Each AI coding agent gets its own isolated machine with root access, Docker, and systemd. Built‑in active defense automatically detects and neutralizes threats.
- OpenMemind: Cognitive Memory AI Agents
Self-learning cognitive memory and context engine in Java for 24/7 active AI agents with understanding and SOTA performance.
- MULTI-AGENT FRAMEWORK ON RUST
AutoAgents — a Rust framework for creating, deploying, and coordinating multiple intelligent agents in a single system.
- Local AI Agent for Chats
Free chat agent with local AI. Works without the cloud, preserving privacy and full control over data.
- LMMS ENGINE FOR MULTIMODAL MODELS
lmms‑engine — a lightweight and flexible framework for training universal multimodal models in Python.
- THOUGHTDAG Mind Map Online
Infinite canvas for LLM dialogues: turn conversations into an editable thought graph, where connections are context.
- MANAGE CODE AGENTS FROM PHONE
Mobile client for Codex and Claude — control agents straight from your phone via WebSocket bridge. 782 stars!
- DRIFT - AI CODE ASSISTANT
Drift is an intelligent code base for AI: it remembers patterns, conventions, and solutions, working offline in any IDE via MCP server and CLI.
- DISTILL: LIGHTWEIGHT CODING AGENT WITH TUI
Distill is a lightweight coding agent with a terminal interface, written in Rust. It enables achieving greater results by using significantly fewer tokens when working with LLM models.
- FORMAE INFRASTRUCTURE CODE FROM EXISTING
Formae is open-source IaC that starts from existing running infrastructure, discovers it and versions drift. It returns current code, serves as a Terraform/Pulumi alternative built on Pkl, and includes an MCP server for AI agents.
- Physical AI Agent OPS-Physics
The first AI physicist with agentic behavior, developed by Physical Superintelligence PBC (PSI).
- Free LLM API Keys
Free API keys for GPT-5.4, Claude, DeepSeek, Gemini, Grok — copy, paste, use. Updated 3-5 times a day; no credit card required.
- MAGIC-CONTEXT - LIMITLESS CONTEXT
The project provides cache‑aware infinite context, inter‑session memory, and background history compression for LLMs.
- NUWAX — PLATFORM FOR AI AGENTS
Corporate platform for creating and managing AI agents: knowledge bases, model proxies, memory, and plugins. 780 stars on GitHub.
- CACHE-TO-CACHE: DIRECT LLM EXCHANGE
Implementation of the Cache-to-Cache algorithm for direct semantic exchange between large language models. Enables transferring KV‑cache without deserialization speeding up multi‑agent interaction. Supports integration with any LLM in Python.
- MACROS FOR SWIFT CODABLE
MetaCodable supercharges Codable via macros. 777 stars — automation of serialization without boilerplate.
- DEEPZERO: FIND VULNERABILITIES WHILE YOU SLEEP
An automated framework finds vulnerabilities in Windows drivers by decompiling and analyzing IOCTL with AI agents.
- PURPLE: One‑Click SSH Management
A free SSH manager written in Rust that syncs ~/.ssh/config with 16 cloud providers and manages containers.
- TRINITY: PLATFORM FOR AI AGENTS
Trinity is a self‑hosted platform for running AI agents Claude Code, Codex and Gemini. It provides orchestration, versioning and access‑policy management. Licensed under Apache 2.0 and implemented in Python.
- PDF READER FOR AI AGENTS
Production-ready MCP server for PDF processing: parallel processing 5-10x faster and 94%+ test coverage.
- CLAUDE CODE JAVA: READY INFRASTRUCTURE
claude-code-java — reusable AI development infrastructure for Java, tailored for Claude Code. It includes projects templates, CLI build scripts, and abilities for refactoring, commenting, and unit testing.
- SPECTRA-APP: SPECIFICATION FOR AI AGENTS
Spectra-app is a desktop app and CLI that turn specifications into working code for AI agents. Supports Codex, Claude Code, Cursor, Copilot, Antigravity and Junie, accelerating development via a specification‑driven approach.
- DAPR-AGENTS: DURABLE AI AGENTS
Framework for building autonomous agents with built-in workflow orchestration, state, and telemetry. Ensures resilience and observability in production.
- SoduWork: ENGINE FOR AI AGENTS
AI-native platform for managing agent teams. Supports multi-harness approach, workflow orchestration, and self-hosted deployment.
- ComfyUI nodes Krea 2
The project adds nodes and ready-to-use workflows to ComfyUI for editing images by instruction while preserving identity using LoRA Krea 2 Identity Edit.
- VOICE AI ASSISTANT FOR MACOS
Atticus is a free voice assistant for macOS. Runs on your Claude Pro/Max plan with no API keys. Spawns background agents for big jobs, remembers you and learns shortcuts. Hold ⌃⌥Space and talk.
- Top Agent Skills
Daily updated top‑100 AI agent skills, based on data from skills.sh, ClawHub, Tencent SkillHub, GitHub, X, and over 10 communities. Data available in open CSV format.
- Nucleoid Logic for LLM
Nucleoid — a Rust logic language for building neuro-symbolic AI that learns and reasons alongside LLMs.
- RECALL — Memory for Claude Code
Recall gives Claude Code long-term memory — fully offline. No need to explain the project again in every session.
- Claude Skill Factory
Open toolkit for creating and scaling production-ready Claude skills, code agents, and custom commands.
- DINGO: AI MODEL EVALUATION
Dingo is a tool for comprehensive evaluation of data, model, and AI application quality in Python. It enables measuring reliability and performance of systems.
- LOCAL-RAG: RAG WITHOUT DATA LEAK
Ingest files for RAG with an open-source LLM — everything runs locally, with no third‑party services and no data leaving your network.
- QWEN2API: TURNING QWEN INTO API
The project turns the Qwen web page into a convenient REST API in Python, allowing quick integration of the model into your applications.
- CREATE AI AGENT VIA GAME
The Jumping-Agent platform lets you create AI agents by training them in interactive game scenarios using TypeScript.
- FREE AI CODER
Project bypasses AI‑coder limits, providing extra requests and access to source code. Users get uninterrupted code generation with full source transparency.
- PRISM MULTI-PROFILE BROWSER
Local-first open-source multi-profile fingerprint browser based on Chromium and Electron. Allows creating isolated profiles with unique fingerprints to protect against detection. Supports automation and operation without servers.
- FLAWLESS AI SRE FOR KUBERNETES
Autonomous AI agent that monitors, fixes, and optimizes Kubernetes clusters and cloud infrastructure without human intervention.
- STARWIND UI: UI COMPONENTS
A set of accessible Tailwind components written in TypeScript that work in Astro and React without framework coupling.
- HTML TO MARKDOWN CONVERTER
Fast and CommonMark-compatible HTML to Markdown converter. Created by the Kreuzberg team — the engine for processing 56+ document formats.
- OPENCLIP — FLOATING TEXT ON MAC
OpenClip — a fast open‑source tool for macOS that shows floating actions over selected text. It lets you transform text, run scripts and connect AI with no extra steps.
- HERMESCLAW - TWO AGENTS IN ONE WECHAT
Run Hermes Agent and OpenClaw simultaneously on one WeChat account. 619 stars on GitHub.
- CODESEEK - SMART CODE SEARCH
A Rust CLI for building function call graphs and hybrid semantic code search across 7 languages. Includes ready‑to‑use MCP tools for Claude Code and Codex CLI.
- CLOUD DESKTOP EXTRA LINUX
Unofficial Claude Desktop packages for Linux with automatic updates. Simplify installation and keep the AI assistant up to date.
- OS FOR SELF-EVOLVING AGENTS
Octopus-SI-OS — agentic OS in Rust with OS-level architecture. Implements self-evolving agents via harness approach to superintelligence. 273★ on GitHub.
- BitFun: Software Agent
BitFun — a desktop agent with evolution and memory and applications for coding and PC work.
- CODANNA SMART CODE
Local code intelligence server in Rust, providing AI agents instant access to context and refactoring.
- OLD-CODER: AGENTS WITHOUT READING
Old coder's strategy: test the code by running it in tests, not by reading it. Evidence-based development for AI agents, inspired by Uncle Bob.
- Terminal Text Annotation
A tool for annotating terminal output, documents, and agents' responses in Herdr, with sending feedback directly to the agent.
- CURATED LIST OF JEV TOOLS
727 stars. Tools for Jev — TypeSafe AI's model for typed decisions. Relevant for LLM and robotics.
- AGENTACC: Agent Report
The panel shows AI agent actions: tools, modified files, launched tests, time and tokens. Works locally, without registration or telemetry.
- CLAUDE EQUITY RESEARCH PLUGIN
Claude Code plugin for institutional stock analysis: generates professional buy/sell recommendations with fundamental and technical analysis.
- OVERMIND: PLATFORM FOR AI AGENTS
Overmind — platform for the continuous improvement of AI agents: automated evaluation, fine‑tuning and integration with LLM operations. It allows tracking metrics, running experiments and deploying updates without manual work.
- ANIMATED LINE-ART SVG COMPONENTS
A TypeScript library and Claude skill for creating animated line-art SVG illustrations: figures walk, react to clicks, scenes have a day/night cycle. Everything is drawn in SVG in real time and stays sharp at any scale.
- REVERSE-API-ENGINEER - Website API Agent
Python AI agent automatically analyzes websites and turns them into full-fledged APIs — no more manual parser writing.
- MQ: JQ for Markdown
A Rust utility for querying Markdown files directly from the command line. 758 stars on GitHub.
- ULTRA FAST OCR FOR RECEIPTS
The receipt-ocr project provides a high-accuracy OCR engine in Python for extracting text from receipt images. It supports image preprocessing, recognition of multiple formats, and integration with LLM for further analysis.
- Startup Stress Test
A skill for testing startup resilience: simulates crisis scenarios and evaluates business readiness for pressure.
- HACKER-NEWS-DIGEST - READ FASTER WITH AI
Automatically summarizes fresh Hacker News with ChatGPT so you stay updated without extra reading.
- Claude Code Animation Skills
Six Claude Code skills that generate reel-ready animations—3D cartoon explainers, stick figures, line art, map routes, exploded views, and graphs—without any source footage.
- Desktop Automation
Native CLI utility for AI agents: control any applications via the desktop API with JSON output.
- WORKSHOP: EVALUATING AI AGENTS
Give your coding agent the ability to write and run agent evaluations. 753 ⭐ on GitHub.
- NORNICDB: ULTRAFAST GRAPH-VECTOR DB
A distributed database featuring sub-millisecond HNSW search, graph traversal, and support for Neo4j Bolt/Cypher plus Qdrant gRPC.
- DSH Desktop: One-click launch
Windows client for dsh with built-in Node.js and CLI. Launch with one click and get DeepSeek power right on your desktop.
- MCP Security Server
MCP server with 27 security tools via 21 APIs: CVE, EPSS, CISA KEV, MITRE ATT&CK, Shodan, VirusTotal.
- LANGGRAPH COURSE: LLM AGENTS
Practical course repository for building production-grade LLM agents using LangGraph, Agentic RAG, ReAct, and reflection workflows.
- JAVA SDK FOR DEEPSEEK-R1
Official library to integrate DeepSeek into Java apps. Supports streaming, function calling, and Spring Boot 3 out of the box.
- SPARKVSR: INTERACTIVE VIDEO SUPER-RESOLUTION
A new video super-resolution method via sparse keyframe propagation. ECCV 2026 paper, 675 stars on GitHub.
- SENTIENT OS: Your AI Assistant
A local LLM in Swift analyzes your data and independently offers to perform tasks for you.
- AI MCP Server for Polymarket
Python MCP server enabling Claude to trade on Polymarket with 45 tools, real-time monitoring, and enterprise-grade security.
- claude-chat-exporter - EXPORT CLAUDE CHAT TO MARKDOWN
Export your Claude AI chats to a beautiful Markdown file with one click. Supports formatting and preserves metadata.
- K-DENSE BYOK: AI Scientist
The k-dense-byok project provides a local AI assistant for scientific research, leveraging the skills of a scientific agent. It runs on TypeScript directly on your desktop.
- OXIDETERM — SSH Workspace in One Binary
A local SSH client in Rust with SFTP, port forwarding, and an AI assistant. No Electron, no OpenSSL, no subscription.
- WOLFCHA - AI MAFIA GAME
wolfcha is a mafia game where all participants are controlled by leading LLMs: DeepSeek, Qwen, Gemini, and others.
- AI Engineer Notebooks
A practical collection of framework-free Jupyter notebooks for learning AI engineer skills: model APIs, structured output, tool calling, RAG, evaluation, agents, fine-tuning vs LoRA, prompt security, and LLMOps, all on the free Groq API.
- MCP SERVER FOR AWWWARDS DESIGN
Free MCP server gives AI agents inspiration from Awwwards: screenshots, design DNA, live site capture. Works with Cursor, Claude Code, Codex — no API keys. Mobbin alternative for coding.
- GIVE MEMORY TO AI AGENTS
Give your AI agents persistent memory.
- METAHARNESS: CREATE YOUR AGENT
Platform for creating specialized AI agents with its own CLI, MCP server, memory, and learning cycle.
- FEEDME - Smart RSS Reader
Lightweight RSS reader built with TypeScript featuring AI hints, runs on GitHub Pages, Vercel, Alibaba Cloud, and Docker.
- AI FOR NOVELS AND DRAMA
DSH plugin creates novels and short-form scripts with AI agents. Written in Python, uses Oh Story and Drama Skills.
- Residual Memory Context Compaction
Remory trains a residual memory module to compact long contexts into shorter representations for transformer models.
- STABLE-BASELINES3-CONTRIB: NEW RL ALGORITHMS
The contrib package for Stable-Baselines3 contains experimental reinforcement learning algorithms. It includes GSDE implementations, Gym environment support, and PyTorch integration. Ideal for fast prototyping of new ML ideas.
- AUTOMEM: LONG-TERM MEMORY FOR AI
AutoMem is a vector-graph memory service that provides AI assistants with persistent relational memory for contextual answers.
- UNIFIED LLM-MCP BRIDGE
Proxy server for centralized access to multiple MCP servers via a single HTTP endpoint in Go. Simplifies integration of resource servers into AI pipelines.
- Unified Embodied Agent Framework
Provides tools to construct embodied agents, benchmark their performance, and iteratively refine them using recursive self‑improvement loops.
- CODEAF: FACTORY OF OPEN MODELS!!
CodeAF — open-source factory for creating and managing AI agents in Go. Enables rapid assembling, testing, and deploying the agent fleets for coding and other tasks.
- SCOUT: OPEN BROWSER AGENT
Scout — open-source browser agent from Outcrawl, automatically mirroring their repository. Written in TypeScript, it controls the browser via CDP and interacts with LLMs to perform tasks in natural language.
- Global Bird Sticker Library
Stickers for all 11,064 living bird species with names in 43 languages and structured data.
- NEXTCRM: OPEN CRM ON NEXT.JS
Open-source CRM built with Next.js 16 and React 19, using PostgreSQL, Prisma, and shadcn/ui. It includes project, invoice, document, email management, and AI features.
- CUGA-AGENT — Universal Agent
An open framework for building AI agents with web/API integration support, OpenAPI/MCP, and advanced reasoning modes.
- PEGAINFER: Rust + CUDA Infer
Pure Rust LLM engine with CUDA acceleration, OpenAI API compatible, runs Qwen3 and Kimi‑K2 without PyTorch.
- AI REVERSE ENGINEERING: CASES AND ALTERNATIVES
Curated verified cases of AI agent reverse engineering: REA, Craft Anything, Claude Code, Codex. 585 stars, bilingual list with open alternatives by app. Every case manually checked.
- Claude Auto-Resume
A shell script that monitors Claude CLI usage limits and automatically resumes interrupted tasks after restrictions are lifted.
- PROBE - FAST AI CODE SEARCH
Semantic code search using ripgrep + tree-sitter for large repositories. Speeds up AI assistants with precise context.
- MLX-DSPARK: 4X LLM ACCELERATION
Portable MLX port of DeepSeek DSpark and z‑lab DFlash for accelerating LLM decoding on Apple Silicon up to 4× without quality loss.
- LEXORA AI AGENT
Lexora — an autonomous desktop AI‑agent written in TypeScript, self‑learning and performing real‑world tasks on Arch Linux. It acts as an independent assistant, so capable of pet‑like interaction and autonomous decision‑making.!
- TEXTUS — AI MARKDOWN EDITOR AP
Textus — next‑level Markdown editor with an AI writing assistant. Free and open source for macOS, Windows, Linux. Integrates DeepSeek LLM for generating text directly in the editor.
- CLAUDE CODE FOR ANY DAY
739⭐ An engineering approach to Claude Code: from chatbot to a full‑fledged system with design, reuse, and management.
- WebUI for GPT-image-2
Interface for GPT-image-2 image generation with Codex Responses and OpenAI-compatible API support, including a shared gallery, prompt chips, prompt templates, and queue management.