Repository index
6792 AI, ML and Web3 projects · page 40 of 68
- MCP Server for Cloudflare API
TypeScript implementation of an MCP server for interacting with the Cloudflare API — convenient integration with your AI agents.
- OPENACCOUNTANTS: AI Accounting
Over 1,000 verified accounting skills for AI agents, covering 192 jurisdictions and all U.S. states.
- Awesome MLSecOps Resources
A curated collection of tools, articles, and resources on machine learning security and MLOps.
- TOOL LEARNING SURVEY FOR LLM
Repository with comprehensive Tool Learning survey: taxonomy, benchmarks, and LLM tool integration methods. 493 GitHub stars.
- Agency: Minimal Framework for Agents
Fast and minimalistic framework in Python for building agent systems. 486 stars on GitHub.
- ANTIGRAVITY IN CLAUDE CODE
Claude Code plugin runs Antigravity CLI as a sub-agent with smart model routing across all SDLC stages.
- LLM2Jev:DECISION MODELS FROM TEXT
The LLM2Jev project converts the local language models into the structured decision models of the Jev style. It outputs the results from the text and images solely via prefill without token‑by‑token decoding.
- K8SGPT Operator: SRE in Kubernetes
The k8sgpt operator adds automated SRE capabilities directly to your Kubernetes cluster, simplifying diagnosis and fixing of issues.
- CHATTS LLM FOR TIME SERIES
ChatTS — the first LLM trained to understand and reason about time series, presented at VLDB'25.
- MAESTRO ORCHESTRATE: MULTI-AGENT PLATFORM
The platform orchestrates 39 AI agents for Gemini CLI, Claude Code, Codex and Qwen Code. It provides parallel sub‑agents, persistent sessions and built‑in tools for code review, debugging, security, SEO, accessibility and compliance.
- Procedural WebGPU Underwater Scene
A fully code-generated underwater world with kelp, shipwreck, and manta rays rendered in WebGPU — all in a single 97 KB HTML file.
- CAD SYSTEM WITH AGENTS FOR CADQUERY
Desktop app for creating parametric 3D models via CadQuery code. An AI agent writes Python code, SOMA builds the model and shows the result in a 3D viewer.
- CLAW ORCHESTRATOR UNIFIED AGENT
Run Claude Code, Codex, Gemini and other CLIs in a unified environment with OpenClaw plugin support.
- VIBE-CHECK: SMART FEEDBACK FOR AGENTS
Vibe Check gives mentor‑style feedback to AI agents, saving them from tunnel vision and over‑complicated workflows.
- INSPIRING INTEGRATION EXAMPLES
A collection of practical examples using Apache Camel to build integration solutions in Java.
- Visual Build Agent Skill
An agent skill that iteratively builds visual outputs to match reference images or videos within a quality threshold and time limit.
- Voice Control Cloud Code
Freestyle — a voice interface for controlling Claude Code, enabling coding without a keyboard. It supports speech recognition and instant command execution.
- LLMFEEDER: One-click LLM
The browser extension turns any web page into clean Markdown and copies it to the clipboard — perfect for quickly feeding context to your LLM.
- Development Sandbox
Self‑hosted dev sandboxes with instant URL previews, launched with a single command. No Kubernetes – ideal for AI agents and SaaS factories.
- ALPHASIFT - AI Engine for Stocks
A full-featured LLM-based engine for stock screening with ranking, risk assessment, and auditable analytics. 219 stars on GitHub.
- AUTOMATING SCIENTIFIC PAPERS IN LATEX
An agentic framework for creating arXiv-ready ML/AI papers with automatic citation checking and a controlled LaTeX workflow.
- Production OCR on Rust
The project demonstrates building an industrial OCR pipeline in Rust, using vLLM for inference, Redis for queuing, KEDA for autoscaling, and Kubernetes for orchestration.
- R3BL OPEN CORE TUI IN RUST
TUI framework in Rust with ready-made components:editor, file manager and process manager. Supports themes, plugins and cross-platform builds.
- VLLM ON C++: POWERFUL SOLUTION
vllm.cpp is an open-source C++ engine, fully compatible with vLLM: continuous batching, paged KV-cache, GGUF support, RadixAttention and cache-aware scheduling. 447★, built by the community for fast LLM inference.
- Grounded Reasoning for Autonomous Driving
GRAVA provides a unified framework that links perception, reasoning, and action for autonomous driving systems.
- UNIVERSAL AI AGENT ON BEDROCK
bedrock-engineer — universal AI agent on Amazon Bedrock. Creates and edits files, executes commands, searches the web, uses knowledge base and multi-agents. Supports image generation, Claude and DeepSeek-R1 models.
- SIGMAP: 97% Token Reduction
Sigmap reduces token usage in AI coding sessions by 97% without external dependencies, supporting 31 languages and featuring a built-in MCP server.
- PR-COCKPIT: LOCAL AI REVIEWER FOR PR
Local AI reviewer powered by Claude and Codex. Reviews code in isolated worktrees, GitHub comments appear only after your approval.
- RUST CLIENT FOR OPENAI API
Library for Rust, enabling work with OpenAI API (unofficial). Supports GPT‑3.5‑Turbo, GPT‑4, GPT‑4o and DeepSeek models. Has 486 stars on GitHub.
- CORDUM - Manage AI Agents
Cordum is an open control plane for AI agents: pre‑execution policy, approval gates, and audit trail. Compatible with LangChain, CrewAI, MCP.
- SKILL.COLOR-EXPERT — COLOR EXPERT FOR CODE
AI agent with deep knowledge of color spaces, accessibility (APCA, WCAG), palette generation, pigment mixing, and the history of color theory.
- MIRA: AI AGENT WITH LONG-TERM MEMORY
Mira — a synthetic entity with a set of primitives for long-term state persistence. The package includes ready‑made modules for an AI‑agent, assistant, and chatbot, including integration with Anthropic Claude.
- Detto: No-Leak Meetings
Detto is a macOS app for recording meetings, dictation, and a local knowledge base without data leaks.
- SWARM: SWIFT AGENT ORCHESTRATION
Swarm — lightweight framework for orchestrating AI agents in pure Swift 6.2, providing scalability and easy integration.
- COMMAND CODE FOR DEEPSEEK HARNESS
The Command Code provider plugin for DeepSeek Harness (dsh) offers access to the Command Code model, a live model catalog, and plan‑aware model selection. It supports adjusting reasoning effort, image input, web search, and multi‑account.
- SOCIAL COACH FOR EQ
SocialCoach — your personal EQ coach. Through role‑play with AI and instant feedback you practice communication and conflict resolution. Works using TypeScript and LLM.
- SKILLS INTELLIGENT VISUALIZATION
skills — intelligent data visualization system based on LLM. Converts natural language queries into ready interactive charts without code.
- AGENT SWARM INTELLIGENCE GROWTH
Agent Swarm – a framework for creating self‑improving agents in TypeScript that accumulates intelligence every day.
- FLYTO-CORE: AI AGENT ENGINE
Open-source engine for executing AI agents with 412 modules, native MCP support, queues, versioning, and usage measurement.
- SKILLALCHEMY: TURN THOUGHT INTO SKILL
SkillAlchemy turns ideas and data into Python skills. Quickly turn thoughts into working code.
- Hand‑drawn explainer video
The project creates hand‑drawn style animated videos to explain concepts, featuring bilingual subtitles/voice‑over and programmable character animation.
- AZURE AI TRAVEL AGENTS
Sample enterprise app deployed on Azure Container Apps, using MCP and multiple AI agents orchestrated by Langchain.js, LlamaIndex.TS and Microsoft Agent Framework. Agents automate flight and hotel search and build routes in real time.
- GERBIL: Local LLMs on Your PC
Desktop app for running large language models right on your computer. No cloud, no subscriptions — full control over AI.
- DIRECTOR — Manage Agents Easily
Platform for creating and managing MCP playbooks for AI agents in TypeScript. Simplifies orchestration and scaling.
- Drissionpage-MCP-Server - Browser Automation for Clients
The DrissionPage MCP server provides browser automation for Claude Code, Codex, and other MCP clients. It allows controlling web pages from Python.
- DB-MCP Multi-DB Server
The db-mcp-server unifies multiple databases via the Model Context Protocol, providing AI assistants with structured data access.
- COMPLETE LLM INFERENCE GUIDE
A collection of best practices, tools, and examples for efficient inference of large language models using TypeScript.
- Rust Wayland Shell Framework
A Rust framework that lets developers build Wayland desktop shells with safe, high‑level APIs.
- UNIVERSAL IOS AUTOMATION WITH MCP
Swipt — iOS automation platform in Go with MCP support. Controls device, apps and system via unified API. Perfect for AI agents and testing.
- 3000+ Bug Bounty Programs
Catalog of active bug bounty and VDP programs with MCP and API. Everything security researchers need.
- KUBESHARK DEFEATS HALLUCINATIONS
This skill for Claude Code and Codex eliminates LLM hallucinations when working with Kubernetes and Helm. LLMs no longer lie — only official best practices.
- ALETHEORE AUDIT REPOSITORY CLI
Aletheore — CLI scanner that deterministically checks the repository, starts an MCP server, shows a live dashboard, and via GitHub Action posts PR diffs.
- A-share data toolkit
A complete toolkit for working with A-share market stock data: 6‑level architecture, 15 endpoints, and 7 data sources.
- MateClaw — Your Digital Brain
A multi-agent AI platform featuring orchestration, memory, skills, and multi-channel support built on Spring AI Alibaba.
- IRONCURTAIN - Smart AI Protection
A secure execution environment for autonomous AI agents using plain‑English policies.
- Voice Dictation for macOS
FrigadeHQ/yap is a free, open‑source voice dictation app for macOS. It performs transcription entirely on‑device using the Apple Speech framework, requiring no internet, API keys, or account.
- OPEN SOURCE SWEDISH ACCOUNTING FOR AI
Accounted — open-source Swedish accounting on TypeScript for people and AI agents. It keeps double-entry bookkeeping, generates reports according to Swedish standards and provides an API for automation.
- LOWFAT - TRIMS COMMAND NOISE
Rust utility that removes noise from command output and saves tokens. Ideal for AI agents and developers.
- CLAUDE CODE MODS: PLUGINS FOR HOOKS
Claude Code mods — a set of hook‑based plugins that add a live line above the prompt, guards, panels and mini‑games. Includes a context bar, usage meter, Codex review watch, Markdown preview and Spotify now playing.
- USE CHATGPT SUBSCRIPTION IN DSH
Connect your ChatGPT Plus/Pro subscription to the DeepSeek Harness to access the GPT‑6 and Codex models,image generation, web search, and quota without a separate OpenAI key. In beta, you can control DSH's from the official ChatGPT mobile app.!!!
- PROCEDURA: TEXT TO 3D MODEL
Procedura converts a text prompt into an editable parametric program for 3D modeling. Supports per‑part materials and articulation of moving parts. Implemented in TypeScript using OpenUSD.
- SELF-HOSTED MAIL SERVER ON DOCKER
Ready-to-use Docker image for your own mail server with IMAP, SMTP, anti-spam, webmail, and admin panel. Deploys in one command, runs on Alpine Linux.
- Phosphene — Live Wallpapers
Video wallpapers for macOS with AI‑enhanced video processing on the desktop.
- AI Agent Research Infrastructure
Platform for automating scientific research with AI agent support: search, analysis, and knowledge generation.
- ISLAND FOR AI CODING
Dynamic Island on macOS for AI agents: monitoring, confirmation, and switching between Claude Code sessions straight from the notch.
- AI-VIDEO-GENERATOR-CLAUDE - VIRAL AI-VIDEOS
10 Claude skills create studio AI prompts for Seedance 2.0 on Higgsfield: viral hooks, SaaS demo, personal brand, and more.
- SOURCEWEFT: OPEN AI WORKING STATION....
SourceWeft — open-source self-hosted alternative to ChatGPT and NotebookLM with agents, skills, MCP and a cited knowledge base. Use your models and infrastructure, getting free access to GPT‑6.
- WEBSITE RECONSTRUCTION BENCHMARK FOR AGENTS
Benchmark for evaluating the ability of AI agents to reconstruct full-featured websites through browser exploration. Contains 70 tasks with 46,916 hidden test cases for verifying visual and functional fidelity.
- LINGHUN - Artificial Intelligence for Code
Linghun is an AGI-oriented code execution environment, resistant to hallucinations, based on evidence, tools, memory, and verification.
- PUPPETMASTER - AGENT SWARM MANAGEMENT
Puppetmaster provides a management plan for resilient agent swarms: subprocesses, leases, artifacts, and deterministic state stitching.
- Hierarchical RAG search
Alpha Agentic Search — a RAG system with hierarchical memory storing knowledge in layers for fast, accurate answers.
- KMCP: CLI AND K8s FOR MCP
kmcp is a CLI utility and Kubernetes controller that allows building, testing, and deploying MCP servers. Integrates with Go projects and supports declarative deployment in the cluster.
- AI-COLLAB-PLAYBOOK - Practical Guide
Practical playbook for collaborative AI work: research, writing, reading, and coding with prompts and agent rules.
- MASTERING LANGGRAPH 1.0
Complete tutorial for LangGraph 1.0 in Jupyter Notebook format. Step-by-step guide for creating AI agents and multi-agent systems.
- WORKER-VLLM: POWERFUL LLM DEPLOYMENT
Runpod worker template for deploying LLM via vLLM — quick start and scalable operation.
- Acoustic Drone Detector
Detects drones by listening for propeller sounds using a microphone, enabling detection of silent fiber‑optic FPV aircraft that emit no RF signal.
- MAIN AI-ASSISTANT FOR CODE
Chief splits work into tasks and sequentially runs Claude Code until each is finished. This allows building large projects with minimal manual effort. Implemented in Go.
- Event-Sourced Actor Framework
Build resilient TypeScript applications using composable actors backed by an immutable event log for state reconstruction and replay.
- ROAM-CODE - AI AGENT FOR CODE
Local CLI + MCP server for AI agents: code graph in SQLite, 28 languages, 238 commands, 224 MCP tools, no API keys required.
- NEXT-GEN JAVA AI FRAMEWORK
solon-ai is a powerful Java framework for building AI applications with LLM, RAG, MCP, and agent support. Compatible with Java 8–25 and SpringBoot.
- DESIGN-STUDIO: AI DESIGNER WITHOUT CODE
Open-source no-code builder on TypeScript, where LLM/GPT meets a flexible graph for fast dialogues and automations.
- FPGA Inference Chip
The project demonstrates the implementation of one transformer layer of Qwen2.5-0.5B in RTL on FPGA, achieving bit‑exact match with the golden model. Measured performance is 0.56 token per second — a 140× speedup over baseline.
- CLAUDE CODE COURSE FOR BEGINNERS
Complete course on AI development with Claude Code. 468 stars on GitHub. From basics to advanced agentic coding techniques.
- HUD FOR 25+ CODING AGENTS
agent-hud-open shows quotas, tokens and sessions for Claude Code, Codex and 25+ agents in one view. Answer requests and jump back to coding in one click. Works on Mac, Windows, Linux, phone and watch.
- PARSEBENCH: AI AGENT EVALUATION
ParseBench is a benchmark for evaluating the quality of document parsing by AI agents. It allows comparing models on real-world datasets.
- SPONSIO - AI AGENT SECURITY
Sponsio offers deterministic security solutions for probabilistic AI agents, minimizing risks of unpredictable behavior.
- Shard: Distributed Inference
Shard implements pipeline parallelism for running large language models across multiple GPU nodes, allowing inference without needing to place the entire model in the memory of a single device.
- PILOT - Accelerate Development with AI
Pilot — the first terminal benchmark 2.0 that uses AI to automatically perform tasks from tickets. It speeds up code delivery and reduces manual work.
- BLOCKRUN-MCP: LIVE DATA FOR AI
Search, research, markets, crypto, Twitter — all via AI agents. Pay-per-use via x402 micropayments.
- PENTEST HARNESS FOR HACKERS
Pentest Harness — a self‑hosted AI‑agent framework for authorized pentests, bug bounty, security labs, and CTFs. Connect your own model via API; all sessions run strictly locally.
- GPU-powered GUI framework
An experimental Rust framework that renders GUI elements directly on the GPU without using a webview, enabling high-performance, low-latency interfaces.
- HCLI WEBSITES READY FOR AI
Hycli turns website tasks into verified CLI and MCP actions, providing a local dashboard and support for 20 languages. Written in Rust, it focuses on local automation of AI agents.
- Buffett Skills for AI
AI skills for investment analysis using Warren Buffett's methodology. Just request an analysis to get a company valuation based on value investing principles.
- VectorRAG.Net — Vector RAG for .NET
High-performance library for semantic search and RAG on .NET. LSH + exact re-ranking via dot/cosine.
- OBSIDIAN + AI: READY!
High-performance MCP server for Obsidian with direct access to the vault via semantic operations and HTTP transport.
- AI FOR BUSINESS DECISIONS
AI agent and data analytics platform for corporations. Helps make business decisions based on AI.
- ONE PERSON COMPANY: AGENT
Platform for creating your own AI-agent company on Python. Easily scale workflow and automate tasks.
- WRITING BENCHMARK – LOVED ELEMENTS
Tests how AI weaves 10 mandatory plot elements into a short story.
- 3D Architecture Atlas
Transforms system architecture discussion into an interactive isometric atlas and generates a SYSTEM.md file.
- ASER: SELF-ASSEMBLY AI AGENT
Aser — lightweight framework for self-assembling AI agents on Python. Supports the a2a protocol, agente memory, and fast deployment via simple scripts.