Trending AI projects

A curated collection of high-fidelity prompts and techniques for Seedance 2.0, helping users create consistent, cinematic, and professional AI-generated videos.

A tool that brings VS Code into the terminal by combining a terminal-based browser and code-server, allowing for a full IDE experience within a CLI environment.

A system that turns an Obsidian vault into a persistent, agent-readable digital brain, allowing AI agents to store, compile, and retrieve knowledge across sessions without relying on external databases.

A self-modifying terminal agent built in Common Lisp that can introspect and modify its own code and state to adapt to tasks.

A plugin that lets senior AI coding agents delegate tasks to Google's Antigravity CLI, providing a fast lane for research, reviews, and implementation using Gemini 3.8 Flash.

Binance Skills Hub is an open marketplace that provides AI agents with the tools to interact with crypto markets, DeFi protocols, and wallet data using natural language.

A text-to-speech conversion tool that turns ePub, PDF, and text files into high-quality audio with synced subtitles using the Kokoro-82M model.

Harvey LAB is an open-source benchmarking suite that provides a dataset of legal tasks and an execution harness to evaluate how well AI agents perform in realistic legal environments.

A marketplace of 22 safety‑and‑automation plugins (hooks) for Claude Code, letting you block dangerous commands, protect secrets, audit sessions, auto‑format code, and more, all installed via simple Claude commands.

Figma MCP Bridge is an open‑source bridge that lets AI agents talk to the Figma desktop app via a persistent WebSocket, bypassing the free‑tier API limit. It provides a rich set of read‑only and optional write tools (e.g., get_document, create_frame, set_text_content) and supports multiple files simultaneously. Install the plugin in Figma, add the server to an AI tool’s MCP config, and start prompting the AI to read or edit design files in real time.

insane-search is a Claude Code plugin that automatically fetches public web pages—even those blocked by 403, CAPTCHAs or WAFs—using a multi‑phase escalation (public APIs → TLS‑impersonated curl → headless browser). It requires no API keys, installs needed helpers on first use, supports many platforms (Reddit, X, YouTube, etc.), can extract PDFs and YouTube captions, and respects login/paywall boundaries. Install via the gptaku‑plugins marketplace and ask Claude as usual; the plugin kicks in when a fetch fails.

TornadoVM is an open‑source Java‑to‑GPU runtime that JIT‑compiles Java bytecode into CUDA, OpenCL, or Metal, offering three Java APIs (@Parallel loops, KernelContext, TileContext) plus native integration with NVIDIA libraries (cuBLAS, cuFFT, cuDNN, cuDF) and Tensor‑Core intrinsics. It runs on NVIDIA, AMD, Intel, Apple Silicon GPUs and CPUs, is installable via SDKMAN!, and is Apache 2.0/GPLv2‑with‑Classpath‑Exception licensed.

A technical book providing a deep architectural analysis of the Agent Harness, the runtime framework that powers AI agents, covering everything from conversation loops and tool systems to memory management and multi-agent orchestration.

A self-hostable dashboard template that acts as a mission control center for the Hermes AI agent, allowing users to dispatch tasks and manage memory via a shared Postgres database.

A Claude Code skill that integrates the latest OWASP security standards (2025-2026) to help developers perform accurate security reviews and build secure AI agent systems.

Bisheng is an open‑source, enterprise‑focused platform that lets teams build and run LLM‑powered applications. It provides a multi‑step “task‑mode” agent, a searchable knowledge base, automated news channels, a visual workflow engine (with loops, parallel branches and human‑in‑the‑loop), and extensive admin features such as RBAC, approvals and audit logs. Installable via Docker, it bundles Elasticsearch, Milvus and OnlyOffice, and reuses LangChain, Langflow, Unstructured and LLaMA‑Factory.

Jean is a desktop‑or‑server workspace that lets developers run multiple AI coding agents (Claude, Codex, Grok, etc.) in parallel, each inside its own git worktree. It bundles project/worktree management, session tracking, built‑in terminal, diff viewer, and deep GitHub/Linear/Sentry integrations, with a web UI for remote access.

A system for decoding natural sentences from non-invasive brain recordings, enabling the reconstruction of text from brain activity during typing.

A free browser extension and bridge that turns popular AI chat interfaces into agents capable of controlling Roblox Studio to write scripts and generate assets.

A comprehensive collection of ComfyUI nodes for end-to-end AI video production, featuring a dedicated scene-based builder, face repair, LoRA training, and audio-to-video synchronization tools.

PhotoPrism is an open‑source, self‑hosted web app that uses on‑device AI to automatically tag, search, and organize personal photos and videos while keeping all data private.

UseAgent is an open-source, self-hostable platform that provides a secure, unified workspace for running various AI agents, featuring isolated sandboxes, persistent event logging, and team-wide tool integration.

ComfyUI‑KJNodes is a plug‑in for the visual Stable Diffusion UI (ComfyUI) that adds utility nodes, cross‑sub‑graph Set/Get handling, and many keyboard/mouse shortcuts to make large generation graphs easier to build and maintain.

A multi-agent collaboration framework that uses an ancient Chinese imperial administrative structure to provide institutional oversight, real-time observability, and strict quality control for AI tasks.

A PyTorch-native stack for running AI models locally on edge devices, from smartphones and laptops to microcontrollers, with optimized hardware acceleration.

iii is a unified runtime and integration framework that simplifies backend service composition, observability, and agentic tool use by mapping all services to a common set of workers, triggers, and functions.

A collection of portable instruction bundles that teach AI coding assistants how to build WordPress using modern best practices, security standards, and Gutenberg patterns.

A canvas-based media workbench for ComfyUI that integrates AI generation with professional-grade editing tools for image, video, audio, and 3D assets.

KServe is a standardized distributed inference platform for Kubernetes that enables scalable deployment of both generative and predictive AI models across multiple frameworks.

LeKiwi is a low-cost mobile manipulator featuring a holonomic drive and a robotic arm, designed for accessible robotics research and data collection.

A curated collection of over 500 data science and analysis resources, including tools, libraries, learning roadmaps, and interview guides.

A Chinese-language, MIT-licensed course that teaches AI engineering by building each algorithm from scratch, with 542 lessons in 20 phases, reusable outputs, and AI-tutor support.

Anthropic's educational repo with five courses (API fundamentals, prompt engineering, real-world prompting, prompt evaluations, tool use) that teach building with Claude, using the low-cost Claude 3 Haiku model to keep student API costs down.

A feed-forward 3D Gaussian reconstruction framework that generates 3D scenes from either a single RGB image or multiple overlapping images for novel-view rendering.

An AI-powered e-commerce customer service SaaS platform that unifies multiple shopping platforms into one interface and automates responses using customizable LLMs.

A production-ready Java framework for building agentic and multi-agent applications, featuring a graph-based runtime for stateful workflows and a visual management platform.

A real-time translation and language learning Android app that uses OCR to translate text from games, visual novels, and manga directly on the screen.

A Burp Suite extension that integrates the security tool with AI clients via the Model Context Protocol (MCP), allowing AI assistants to interact with Burp Suite.

Stirrup is a lightweight Python framework for building LLM‑driven agents. It supplies ready‑made tools (code execution, web search, file I/O, multimodal handling) and a flexible `Tool`/`ToolProvider` system, while automatically managing context limits and session lifecycles. Install via `pip install stirrup` (optional extras for Docker, browsers, etc.), create a client (OpenRouter, LiteLLM, or any OpenAI‑compatible API), instantiate an `Agent`, and run a session that lets the model invoke tools to solve tasks. Custom tools and providers are easy to add, making Stirrup suitable for rapid prototyping, domain‑specific assistants, and research on tool‑using agents.

The official front-end for ComfyUI, providing a visual node-based interface and a robust extension API for designing AI workflows.

A cross-platform 3D reconstruction engine using Gaussian splatting that runs on WebGPU and the Burn framework to support macOS, Windows, Linux, Android, and browsers.

A desktop GUI for the Pi and oh-my-pi coding agents that integrates chat, project management, and a code editor into a single window.

A graphical user interface for AI-powered image and video upscaling and frame interpolation, supporting a wide range of hardware and neural network models.

timm (PyTorch Image Models) is a library that provides hundreds of pre‑trained vision architectures (CNNs, Vision Transformers, hybrids, etc.) together with training scripts, optimizers, schedulers, data utilities, and hyper‑parameter sweep tools, all built on PyTorch for easy experimentation and deployment.

tweakcc is a Node‑based CLI that customises the Claude Code AI‑assistant desktop app. It patches Claude Code’s JavaScript (or native Bun binary) to let users change system prompts, UI themes, spinner animations, toolsets, model aliases, session handling, and performance settings. Configurations are stored locally (or fetched remotely) and can be applied via an interactive UI, command‑line flags, or programmatically through its npm API.

A collection of tutorials and reproducible code for state-of-the-art Agentic Reinforcement Learning algorithms, designed to lower the barrier to entry for researchers and developers.

A repo of MCP servers that let AI assistants read, analyze, and modify Cloudflare account configurations via natural language. Includes one recommended broad-access API server (in a separate repo) plus domain-specific servers for documentation, containers, browser rendering, monitoring, security, and more. Each server exposes a stateless HTTP handler at /mcp; the old /sse endpoint returns a 410 error. Connect via any remote-capable MCP client or Cloudflare AI Playground. Some features require a paid Workers plan.

A collection of ready-to-use Google Colab notebooks for fine-tuning, training, and running inference on various open-source LLMs and multimodal models.