Trending AI projects

K-Dense BYOK is an open-source AI research assistant that runs locally, helping scientists analyze data, review literature, and generate reports while maintaining a transparent, verifiable lab notebook.

Newsjack is an open‑source toolkit that turns LLM agents into a PR team. It provides a set of Markdown‑based “skills” (e.g., angle generation, fact‑checking, journalist finding) and a small CLI, installable via a one‑line script or as a Claude.ai plugin. The skills run fully on local agents (Claude Code, Codex, Hermes, OpenClaw) and in limited one‑shot mode on Claude.ai and ChatGPT. It targets founders, agencies, and marketers who want AI‑assisted news‑jacking, monitoring, and PR workflow automation.

An MCP server providing fast, cheap, typed judgment tools from TypeSafe's Jev model to help AI agents perform high-frequency verification and screening tasks.

Brigade is a self‑hosted, multi‑agent AI assistant framework. It lets you run a crew of agents that share persistent, provenance‑aware memory (Tideline), are organized in an org‑chart, and can be reached via terminal UI, WhatsApp, Slack, Discord, iMessage, etc. All API keys and data stay on your machine; you can bring any LLM (Claude, OpenAI, Gemini, local Ollama, etc.) and connect to 1 000+ third‑party apps through Composio. Features include skills, sub‑agent fan‑out, cron scheduling, document/media processing, optional Convex storage, and an MCP memory server. Install with a one‑line script or via npm, then onboard, start the always‑on gateway, and chat.

zerostack is a lightweight coding agent written in Rust that runs in the terminal, supporting multiple LLM providers. It offers granular permission controls, session management, subagents, and optional persistent memory, while using far less RAM and CPU than JavaScript-based agents.

A collection of 31 AI agent skills for film and TV script creation, covering everything from novel screening and story analysis to script generation and professional evaluation.

Qualcomm AI Hub Models is a Python package and CLI that provides a catalog of pre‑optimised machine‑learning models (vision, language, generative AI) ready for deployment on Snapdragon CPUs, GPUs, and NPUs. It handles quantisation, compilation, and performance profiling on real Qualcomm devices via the AI Hub Workbench, and includes sample apps and demos for both local PyTorch inference and on‑device accelerated runtimes.

A multimodal toolbox for evaluating embeddings and retrieval systems, providing a standardized benchmark and leaderboard for comparing model performance.

wonderful-prompts 是一个活跃维护的中文 ChatGPT 提示词精选集，收录上百个高质量提示词，覆盖写作、编程、绘画、生活、职业等广泛场景，并配有图文示例。项目由 ChatGPT 中文指南作者创建，MIT 许可，持续接受社区贡献，并附有面向 GPT-5、Claude、DeepSeek-V4 等当代模型的提示技巧说明。

A Linux-based voice satellite for Home Assistant that runs on x64 or ARM64 hardware, enabling custom smart speakers with local wake word detection and seamless integration via the ESPHome protocol. It handles on-device wake word recognition and provides flexibility for additional features, all while connecting to Home Assistant's Assist voice system.

A modern POSIX shell toolkit that provides a standard library of 300+ modules and 600+ on-demand CLI tools, specifically optimized for AI agents to perform system operations.

X‑VLA is a transformer model that unifies vision, language, and robot proprioception to predict actions across many robot types. It uses learnable soft prompts to specialize a shared backbone for each embodiment, runs as a REST inference server, and provides many pre‑trained checkpoints (foundation and robot‑specific) on Hugging Face. The repo includes installation scripts, server/client code, and instructions for fine‑tuning on custom robot datasets.

A plugin for AI coding agents that provides a relational knowledge graph and a library of specialized skills to manage and deploy Vercel projects.

A self-hosted AI accounting app that automates expense and income tracking by extracting data from receipts and invoices using LLMs.

nnsight is a Python library for tracing, reading, and editing the internal tensors of any PyTorch model (including Hugging Face Transformers, Diffusers, and vLLM). It lets you intervene in a model’s forward pass with simple Python syntax, supports gradient capture, generation‑step hooks, batch‑wise prompts, and remote execution via NDIF, making it useful for interpretability research and custom LLM tooling.

A human video matting framework that preserves fine details and avoids coarse boundaries to extract people from videos with high robustness in real-world conditions.

JADX‑MCP‑SERVER is a Python server that lets LLMs (like Claude) talk to the JADX decompiler via the Model Context Protocol. It exposes many reverse‑engineering actions (fetch class source, search, rename, add comments, debug info, etc.) as “tools” that an LLM can call, and includes an OpenAI‑compatible client for easy integration.

Polymarket MCP Server is a Python‑based MCP server that lets Claude (or any LLM) autonomously discover, analyse, trade, and monitor Polymarket prediction‑market positions. It offers 45 tools across market discovery, analysis, trading, portfolio management, and real‑time monitoring, all guarded by configurable safety limits and full‑node authentication. A web dashboard provides visual control, and the project includes real‑API tests, MIT licensing, and detailed docs for both demo (read‑only) and live‑trading setups.

A command center for managing multiple AI agents, featuring a Kanban dashboard, automatic Git Worktree isolation for conflict-free coding, and multi-agent collaboration.

A reconstruction of Joseph Weizenbaum's original 1965 ELIZA chatbot, running on an emulated IBM 7094 computer and the Compatible Time-Sharing System (CTSS).

An open-source auto-aiming and fire-control system for combat robots that uses neural network detection and trajectory planning to maximize hit rates and minimize kill times.

Open‑XiaoAI Bridge is a server‑side bridge that lets a Xiaomi XiaoAI speaker talk to external AI services (OpenAI‑compatible APIs, OpenClaw agents, QwenPaw, XiaoZhi). It provides wake‑word detection, VAD, optional offline ASR, multi‑agent routing, continuous dialogue, and a Docker‑ready HTTP API for remote playback. Configuration is done via `config.py` and environment variables; TTS can use the speaker’s native engine, Doubao cloud, OpenAI TTS, or a local MLX‑Audio model.

AAS is an open‑source, Docker‑based framework for multi‑drone development, fast ROS 2 simulation (PX4/ArduPilot, Gazebo, YOLO/LiDAR), and deployment on Jetson hardware, with optional RL Gymnasium support.

An open-source gallery of web UI experiments and components generated with Claude, providing the code and prompts for landing pages, 3D scenes, and animations.

Markstream is a family of streaming Markdown renderers designed for AI chat and real-time data, preventing visual flicker and layout instability while content is still being generated.

A zero-config documentation framework that turns Markdown/MDX folders into AI-ready documentation sites with built-in support for MCP servers and agentic tooling.

ABot‑Recon is a research‑grade system that reconstructs dense 3‑D maps from long video streams while only ever keeping a 12‑frame local context. It predicts per‑frame point clouds and relative poses, composes them into a global trajectory, and runs at ~24 FPS on an H100. The repo provides installation scripts, a Python API, optional loop‑closure, visualization tools, and pre‑trained checkpoints on Hugging Face.

A curated npm package that bundles hand-picked agent skills from across the ecosystem, so projects get them installed and auto-synced with a single dependency.

A Chinese-language, 37-chapter hands-on course that teaches AI agent development from fundamentals to advanced production topics. Each chapter is a runnable Python file, and a free live website is provided for browser-based learning.

A Swift package that integrates Claude into Apple's Foundation Models framework, allowing developers to use Claude as a server-side model via native Apple APIs.

A local-first whiteboard and diagramming application that allows users and AI agents to collaborate on an infinite canvas without requiring an internet connection.

A portal that aggregates abstracts from top computer vision, language, machine learning, and robotics conferences into searchable web pages, helping researchers quickly find papers by keyword across many years and venues.

A code-first TypeScript toolkit for building, orchestrating, and deploying AI agents with full type safety and support for both Node.js and browser environments.

This is a free, structured study roadmap for learning AI and machine learning, built as a GitHub repository. It curates links to free courses, videos, and articles across 11 modules, ranked by difficulty, plus a bonus module, and includes extra sections for projects, websites, newsletters, and blogs.

A collection of pure‑Python, from‑scratch implementations of classic ML algorithms, deep‑learning models, and applied projects (recommendation systems, CV apps, NLP, etc.). Designed as a hands‑on learning resource for anyone who wants to understand the inner workings of machine‑learning techniques.

A collection of standardized, expert-level instructions for AI coding assistants to help them generate high-quality, modern Android code consistently. It replaces repetitive manual prompting with an automated, modular skill-loading system.

A community-curated list of AI-assisted coding tools and articles, organized by category — web builders, editors, CLI tools, plugins, and more — centered on the “vibe coding” approach of developing software through prompts and iteration.

A hands-on course that teaches you to build a production-grade, self-scaling OCR pipeline on Kubernetes using Qwen 3.5, vLLM, a Rust gateway, and KEDA autoscaling, with MCP integration for AI agents.

A comprehensive AI Agent workflow and skill pack for professional industry and corporate research, transforming public information into verifiable reports with financial models and competitive analysis.

OpenDM is Dexmal’s open‑source Vision‑Language‑Action model (DM0.5) for robot control, providing pretrained weights, Docker‑based training/inference pipelines, dataset registration, and benchmark‑level performance on tasks like LIBERO, RoboDojo‑Sim and RoboColiseum.

A community-maintained directory (

A hands-on tutorial repo that teaches building AI agents from scratch using local LLMs (via node-llama-cpp), walking through 15 progressive examples from basic LLM calls to advanced reasoning patterns like ReAct, Tree of Thought, and Graph of Thought — all without frameworks.

A lightweight Rust-based proxy that turns Claude.ai and Claude Code into OpenAI-compatible API endpoints using session cookies.

A framework that converts local LLMs and VLMs into structured decision models by extracting probabilities from logits during prefill, bypassing token-by-token decoding for lower latency.

Ludwig is a declarative deep learning framework that enables users to train, fine-tune, and deploy LLMs, multimodal, and tabular AI models using YAML configuration files instead of boilerplate Python code.

CuPy is a library that provides a drop-in, GPU-accelerated alternative to NumPy and SciPy, allowing for faster numerical computing on NVIDIA and AMD hardware.

Yao GEO Skills is an open‑source catalog of 24 reusable “skill” packages that automate the end‑to‑end workflow of Generative Engine Optimization – making web pages and brand content AI‑search‑friendly, generating knowledge graphs, titles, comparison reports, and tracking the impact of those assets. Each skill includes clear documentation, a manifest, templates, validation tests and example outputs (Markdown, HTML, DOCX, PDF). The repo is geared toward marketing, SEO and growth teams that need repeatable, auditable processes for AI‑driven search, and integrates with the GEOFlow platform.

An enterprise security system for AI agents that provides discovery, observability, and threat detection to secure employee-facing and customer-facing AI tools.