Trending AI projects

A macOS sandboxing tool that uses sandbox-exec to restrict LLM coding agents to a least-privilege file system access model.

A collection of opinionated configurations, sandboxing rules, and hooks for Claude Code to enable secure, high-throughput autonomous AI coding.

A cross-platform desktop application that provides a unified, visual interface to manage configurations, sessions, and plugins for various AI coding assistants and CLI tools.

Bug Hunter is an adversarial AI code-audit tool that uses a multi-agent system of hunters, skeptics, and referees to filter out false positives and provide evidence-backed security and behavioral bug reports.

A zero-dependency CLI tool that lets you generate images using your existing ChatGPT or Gemini subscriptions without requiring separate API keys or additional billing.

RhinoAI is an official Rhino 3D plugin that connects the CAD environment to large‑language‑model assistants (Claude, Copilot, OpenAI, Gemini, or local models). It lets users issue natural‑language prompts to generate or edit geometry, offers a one‑click installer, VS Code debugging support, and extensive docs for both end‑users and developers.

PointWorld is a large pre-trained 3D world model that predicts full-scene 3D point flows from partial RGB-D captures and robot actions to improve in-the-wild robotic manipulation.

A high-performance Go-based AI backend for ESP32 devices that enables low-latency, full-streaming voice interactions by integrating ASR, LLM, and TTS engines.

An open-source framework that gives AI agents a mobile body, enabling them to control Android and iOS devices via a unified tool surface and MCP server.

herdrm is a native macOS app that provides a graphical console for the **herdr** runtime, which hosts AI coding agents (Claude, Codex, Gemini, etc.). It aggregates local and SSH‑connected devices into a single sidebar, shows live status of spaces/agents/terminals, and lets you attach to any agent via a real PTY terminal (using the Ghostty engine). Features include file drag‑and‑drop, two‑pane remote file manager, system notifications, searchable Claude conversation history, and universal‑binary distribution. Install via Homebrew or a signed zip; requires macOS 14+, a local/remote **herdr** daemon, and SSH access. The project is early‑stage, open‑source under a non‑commercial PolyForm license, and welcomes contributions.

A high-performance CPU, GPU, and memory profiler for Python that provides line-level analysis and AI-powered optimization suggestions.

A custom mpv video player build and a set of Real-ESRGAN models that enable real-time 4K upscaling of anime videos using TensorRT or DirectML.

A JetBrains IDE plugin that runs a local Model Context Protocol (MCP) server, exposing the IDE’s indexing, refactoring, diagnostics, and build capabilities as JSON‑RPC tools for AI coding assistants (Claude, Codex, Cursor, etc.). Supports multiple languages, headless operation, and optional project‑lifecycle management.

An intelligent agent-powered document search platform that enables users to upload, process, and query documents through a chat interface using RAG.

A high-accuracy speech-to-text system that allows users to choose between verbatim transcription (including fillers) and clean, intended text, featuring precise word-level timings.

SALMONN is an open‑source family of multimodal large language models that jointly handle vision, speech, text and actions. The repo provides model checkpoints, inference code, training scripts, and large annotated datasets (including video‑captioning and speech‑quality data). Variants such as video‑SALMONN‑2, ELLSA, and F‑16 enable streaming, full‑duplex perception and generation, making the models suitable for interactive agents, video QA, captioning, and speech‑quality assessment.

A universal multi-engine TTS and audio suite for ComfyUI that integrates voice synthesis, voice conversion, and SRT subtitle synchronization across dozens of AI models.

SLayer is an embeddable semantic layer that provides AI agents with a governed, flexible interface to query data warehouses, handling SQL translation and complex transformations automatically.

A GitHub Action that integrates Claude Code into CI/CD workflows, enabling automated agentic coding tasks like PR reviews and dependency updates.

An Obsidian plugin that integrates AI agents via the Agent Client Protocol, allowing agents to read and edit notes in your vault without copy-pasting.

A professional data analysis skill for AI agents that prevents silent calculation errors in Excel by implementing a rigorous 8-step audit and reconciliation workflow.

x-reader is a Python CLI and Agent Skill that fetches the true source behind URLs (tweets, videos, articles, etc.) and reports whether the full content was read, partially read, or failed, providing machine‑readable evidence for AI agents.

An MCP server that gives AI agents free, keyless web search and anti-bot fetching capabilities, allowing them to bypass bot walls and crawl the web locally.

Observer AI is a local-first agent framework that monitors your screen, camera, and audio to trigger automated tasks and notifications based on visual or text-based events.

A pure-Python reference implementation of the CuTe layout-and-tensor algebra used in CUTLASS 3.x, enabling layout prototyping and testing without a GPU.

A web-based interface for Music-Source-Separation-Training that allows users to isolate vocals and instruments from music tracks using various AI models.

A TypeScript framework that simplifies building Model Context Protocol (MCP) servers by providing an opinionated, high-level API that handles boilerplate, transport, and authentication.

Video Recap Skills is a plug‑in suite for AI agents that automatically turns any video into a short Chinese‑narrated recap. It uses Xiaomi MiMo for speech‑to‑text, visual‑language analysis, and text‑to‑speech, while only requiring ffmpeg locally. The agent creates a story plan, selects clips, writes narration, generates TTS, mixes everything, and outputs an MP4 plus subtitles. It can also export a multi‑track Jianying/CapCut draft for further manual editing. Installation is a few `brew/apt` commands, setting a MiMo API key, and adding the plug‑in to Claude Code, Codex CLI, OpenCode or OpenClaw. The repository is MIT‑licensed and part of the ZenStory AI open‑source ecosystem.

Mission Planner is a ground control station software for ArduPilot-based unmanned vehicles, providing tools for mission planning, vehicle configuration, and firmware management.

dsh‑pentest is a DeepSeek Harness plugin that adds a structured penetration‑testing mode. It provides tools for agents to record goals, intents, facts, findings and assets, stores them in a SQLite DB, and shows the data in a dedicated web UI (graph, vulnerability list, asset view, markdown report). Intended for authorized security work, it works with DSH ≥ 0.1.6 and Node ≥ 22.5.

A local deployment skill for MiniMax H3 that enables low-VRAM NVIDIA GPUs on Windows to generate videos with native audio via AI agents.

A command-line tool that syncs X bookmarks to a local knowledge base, enabling LLM-powered classification and providing AI agents with grounded local context.

A unified vision foundation model that distills multiple large models like CLIP, DINOv2, and SAM into a single backbone for improved efficiency and versatility across visual domains.

SentrySearch is a command-line tool that indexes video footage into a local vector database, letting you search with natural-language text or a reference image and save the best matching moment as a trimmed clip.

A framework for building multi-agent AI applications that organizes agent collaboration using real-world organizational structures and directional communication flows.

A tool and SDK that lets you use your ChatGPT account as an OpenAI-compatible API, enabling developers to build apps where users bring their own AI accounts.

ProjectAtlas is a Rust-native CLI and MCP server that creates a local, intelligent map of your codebase to help AI coding agents navigate large repositories more efficiently and reduce token costs.

Relax is an open‑source, Ray‑Serve‑based reinforcement‑learning engine for multimodal large language models. It supports text, vision, video, and audio in a single training loop, offers fully asynchronous or hybrid execution across separate GPU clusters, includes many RL algorithms (PPO, GRPO, M2PO, RLOO, REINFORCE++…), and provides production‑grade features such as elastic rollout scaling, health‑checking, and centralized metrics. The framework ships as a Docker image with Megatron‑LM training and SGLang inference back‑ends, and comes with ready‑to‑run scripts for text, vision‑language, and omni‑modal tasks.

An open-source design canvas that integrates AI-powered image and asset generation with traditional graphic design tools for macOS, Windows, and Linux.

Aixora Audio is a Windows‑focused, all‑in‑one desktop UI that runs open‑source speech‑synthesis, transcription, translation, dubbing, real‑time voice chat and other audio tools locally via a compiled C++ inference backend (audio.cpp). It supports CUDA, Vulkan or CPU, offers a large catalog of models, and provides a portable package plus source‑build instructions.

StatsPAI is an open‑source Python library that replicates many Stata and R econometric commands (regression, IV, DiD, RD, synthetic control, matching, etc.) with a Stata‑style API, provides agent‑ready metadata, built‑in datasets, and utilities for exporting tables and interactive plotting.

Ceres Solver is a C++ library for modeling and solving large, complex non-linear least squares and general unconstrained optimization problems.

A C++ and Python library for efficient Transformer model inference, using quantization and custom runtimes to accelerate execution and reduce memory usage on CPU and GPU.

Darts is a Python library that simplifies time‑series forecasting and anomaly detection. It offers a unified `fit`/`predict` API for classic models (ARIMA, Exponential Smoothing) and deep‑learning models (N‑BEATS, TCN, Transformers), supports multivariate series, covariates, probabilistic forecasts, back‑testing, and integrates with PyTorch Lightning and MLflow.

An open-source, self-hosted companion app for AI voice recorders that syncs recordings and transcribes them using user-chosen AI providers.

OpenLLM is a Python tool that lets you run open‑source LLMs (e.g., Llama 3, Qwen 2.5, Mistral) locally and expose them through OpenAI‑compatible REST APIs. With a single command you can start a GPU‑accelerated server, use a built‑in web chat UI, and deploy the service via Docker or Kubernetes. It supports a catalog of models, custom model repositories, and integrates with the high‑performance vLLM backend.

AITuberKit is an open‑source toolkit for building AI‑driven virtual characters (VTubers) that can chat, speak, and stream live. It integrates many LLMs, TTS engines, and avatar formats, and offers modes for YouTube chat, kiosk demos, and game commentary. Run locally, via Docker, or deploy to Vercel/Cloudflare. Commercial use needs a separate license.

A SQLite extension that enables offline-first, conflict-free data synchronization across devices using CRDTs, designed for collaborative apps and distributed AI agent memory.