Trending AI projects

A Next.js‑based editor skill for AI agents that helps developers create and export App Store / Google Play marketing screenshots with themes, connected canvas, and multi‑platform support.

A native visual design system for DeepSeek Harness that allows AI to generate and manually refine editable websites, presentations, and videos.

A multi-agent coding workspace that integrates various AI coding agents into a single visual interface, allowing them to collaborate and work on the same project.

A Python harness for building auditable fact-verification loops that turns claim verification into a replayable evidence circuit with versioned sources and immutable traces.

Ming-3.0-tts is a controllable multilingual text-to-speech system with two models (0.5b dense, tiny MoE), supporting voice cloning/design, free-form instructions, paralinguistic tags, emotion/attribute control, dialects, and multi-speaker dialogue. It ranks 1st among open-weights models on Artificial Analysis's Controlled Voice leaderboard and 1st overall on Pronunciation Robustness. Each capability is a separate task mode; mixing is rejected. The README lists a Limitations section but its content isn't in the provided excerpt.

A physical, 3D-printed robotic companion that connects to your AI coding agents via Wi-Fi to provide visual and auditory feedback on their status and activity.

MatrixMedia is a multi-platform video distribution tool that automates batch uploads to various social media sites, featuring a CLI, HTTP API, and MCP support for integration with AI agents and automation workflows.

An open-source, scalable WebRTC infrastructure for routing real-time voice, video, and data between humans and AI agents.

Pinvou Agent is a cross‑platform desktop AI workspace built with Tauri 2, React and a Rust‑based agent engine (CodeWhale). It lets users attach files, run local or cloud LLMs, and use extensible tools to turn chat prompts into real deliverables—documents, designs, or code—while keeping data private when desired.

A collection of 35 paste-ready prompts and configuration files (system prompts, AGENTS.md files, Ollama Modelfiles) that re-shape model behavior for 25+ AI models, documented for research and education on prompt injection and instruction hierarchy.

Nudgebee is an open-source SRE copilot that automates observability, incident response, and FinOps across Kubernetes and cloud environments using LLM-driven triage and runbook automation.

Easel is an AI-powered social media workbench that automates the entire content creation lifecycle, from trend discovery and multi-format media production to cross-platform publishing and performance analysis.

An Android-based AI conversation co-pilot that analyzes incoming messages in real-time across various chat apps and suggests ranked replies based on a local knowledge base.

TrendRadar is an open‑source, quickly‑deployable news‑hotspot assistant that aggregates headlines from many sources, filters them with user‑defined keywords, and optionally uses LLMs to translate, smart‑filter and generate AI‑driven trend analyses. It pushes concise digests to dozens of chat/notification platforms, offers a web‑based HTML report, supports custom schedules, self‑hosted data sources and a Python MCP client for programmatic use.

jcode is a high-performance framework for running multiple AI coding agents in parallel, featuring native swarm collaboration, autonomous self-improvement, and significantly lower memory overhead than traditional coding assistants.

Token Monitor is a desktop widget that aggregates live token usage, cost, and AI-tool account limits across 46+ coding agents/IDEs by parsing local logs and databases, with real-time multi-device sync via SSE or iCloud.

A graphical desktop management tool that aggregates AI API providers, manages credentials, and configures local proxy settings for various AI agent clients.

VibeWise is a Claude Code / Codex plugin that adds a learning‑first workflow to AI‑generated code. It guides you through design discussions, explains concepts, and only writes code after you confirm the plan, storing all notes locally in a `.vibe-wise/` folder.

A high-performance graphical web browser that runs inside your terminal, enabling developers and AI coding agents to interact with the web without leaving their command-line environment.

Termio is a macOS (and iOS) application that creates a persistent, terminal‑first environment for AI coding agents. Sessions run in a daemon (`termiod`) so they survive app closures and can be accessed locally, via SSH on remote servers, or from a mobile companion. The UI offers project‑based organization, git worktree handling, split panes, a built‑in editor, live diff, and a status indicator that tells you when an agent needs human input. A rich CLI (`termio`) lets you spawn, control, and monitor agent sessions, while the roadmap adds a tmux‑like TUI, Android, Windows, and web clients.

A suite of free AI-powered coding and research tools that provides access to multiple leading LLMs via specialized agents across CLI, desktop, and web interfaces.

A comprehensive library of modular skills and plugins that enable AI agents to interact with, manage, and troubleshoot Google Cloud services and developer tools.

Prime Agent is an open‑source command‑line AI assistant that lets a large language model act as a persistent, programmable coding/research agent. It provides a Python REPL, sub‑agent spawning, a durable “continual harness” for storing prompts, memories and reusable skills, background daemon support for long‑running sessions, and autonomous execution with budgeting and quality‑gate controls. The tool is aimed at researchers and developers who need AI‑driven automation that can survive across terminal sessions.

fframes is a Rust-based framework for creating programmatic videos using SVG and GPU-accelerated rendering, designed to be easily managed by AI coding agents.

An AI learning system built for the pi framework that uses specialized subagents and interactive extensions to provide a structured, visual, and verified learning experience.

An AI roleplay interface built on DeepSeek Harness that improves stability and performance by separating narrative generation from background logic, offering a more immersive and reliable experience for character-based storytelling.

Omnigraph is an open‑source graph database built on the Lance columnar format that runs on any S3‑compatible object store. It offers typed schemas, multimodal retrieval (graph + vector + full‑text), Git‑style branching, and Cedar‑based security policies, all managed via Terraform‑like `cluster` files. Designed for long‑horizon AI agents, it lets multiple agents share a versioned world model and retrieve context in a single query.

Navop is a high-performance, all-in-one developer workspace that integrates database management, remote server access, and AI agent workflows into a single native desktop application.

CloakBrowser is a custom‑patched Chromium binary that removes automation fingerprints at the C++ source level. It provides drop‑in Python, JavaScript, and .NET wrappers that mimic Playwright/Puppeteer, includes a `humanize` flag for realistic mouse/keyboard actions, supports proxies with geo‑IP spoofing, and offers a free single‑session tier plus paid plans for scaling. Designed for web‑scraping and AI agents that need to bypass bot‑detection services like reCAPTCHA, Cloudflare Turnstile, and FingerprintJS.

An open-source enterprise AI data hub that connects fragmented databases to AI applications, enabling natural language data retrieval and automated data governance.

Carrel is a self‑hosted knowledge‑base platform for AI agents. It parses many document types, builds vector/keyword indexes and optional Neo4j knowledge graphs, and exposes a REST API (with a ready‑made search skill) that returns passages, images and graph evidence. Deployable via Docker Compose/systemd, it works with local or remote embedding, vision and chat models.

SenseNova-Skills is a collection of modular, agent-compatible tools that enable AI to perform end-to-end office tasks like data analysis, slide-deck generation, and deep research.

A real-time interactive streaming engine for digital humans that synchronizes audio and video for lifelike AI-driven dialogues.

Kordoc is a powerful document parsing and generation engine that converts complex Korean office formats (HWP, HWPX) and PDFs into structured data or Markdown, enabling AI agents and RAG systems to process them with high layout fidelity.

Falcon-TST is a collection of large-scale foundation models designed for efficient and accurate time series forecasting across univariate and multivariate data.

A high-performance, terminal-based AI coding assistant that provides a lightweight, keyboard-driven interface for managing files, running shell commands, and executing complex coding tasks with LLMs.

A CLI tool for evaluating and evolving AI agent skills through declarative test cases, automated environment provisioning, and an AI-driven feedback loop.

iMCP is a macOS app that exposes your calendar, contacts, messages, HomeKit devices, and other personal data as MCP‑compatible tools for AI clients like Claude Desktop, letting large language models retrieve or act on that information securely and locally.

Beads (bd) is a CLI‑driven, Dolt‑backed issue tracker that gives AI coding agents a structured, version‑controlled task graph and persistent project memory, enabling conflict‑free multi‑agent collaboration.

An open-source, on-device AI agent for iOS and Android that integrates a sandboxed Linux shell and deep system access to perform complex mobile tasks.

yovoice is an open‑source, local‑first desktop studio (macOS 14+ Apple Silicon & Windows 10/11) for turning text into expressive speech, cloning custom voices, and generating full songs and MIDI scores. It bundles the C++‑based audio.cpp engine and a range of TTS/music models (VoxCPM2, IndexTTS, Qwen3‑TTS, ACE‑Step 1.5, etc.), runs entirely on your GPU/CPU, and offers a GUI, CLI, and an authenticated HTTP/MCP API for automation. No cloud calls, no per‑character fees, and full privacy.

A provider-agnostic TypeScript toolkit for building AI-powered applications and agents with a unified API for multiple LLM providers and integrated UI hooks.

AI Security Tool is an open‑source, cross‑platform suite that combines a multi‑protocol terminal (SSH, RDP, VNC, etc.) with AI‑driven security automation. It offers supply‑chain & CVE scanning, LLM‑based command assistance, and a set of 12+ autonomous agents that handle tool selection, recon, attack‑chain building, and fault tolerance. Available as native installers for Windows, macOS, Linux, Android, and as a web UI, it targets red‑teamers, bug‑bounty hunters, and CTF players looking for a unified AI‑native workflow.

A free, company-by-company list of real AI-engineering interview questions from 35 companies, with linked answers where available and common questions grouped by topic; maintained by Outcome School.

An official inference framework for 1-bit (ternary) LLMs that enables fast, energy-efficient, and lossless inference on CPUs and GPUs.

LobeHub is an open‑source platform for orchestrating AI agents as a 24/7 “team”. It lets you create agents via a builder, give them access to any LLM and thousands of plugins, schedule and monitor their work, and store editable personal memory. Deployable via Vercel, Zeabur, Alibaba Cloud, or a Docker image, it requires only an OpenAI API key. The project also provides UI libraries for building AIGC web apps.

EAGLE‑4 is a source‑available library that adds adaptive speculative decoding (a “drafter”) to LLM serving engines (SGLang, vLLM, TokenSpeed). It automatically sizes the speculative tree based on batch size, hardware and workload, delivering up to ~3× token‑per‑second speed‑up on NVIDIA B300 or AMD MI355X GPUs. The repo provides the `eagle_harness` Python package, installation guides, pre‑built wheels/Docker images, and a support matrix with measured results. License is source‑available with commercial‑use limits.

prompt2rig turns text prompts into 3D character animations. It uses a video model to generate a clip, then extracts clean 3D body motion (GVHMR) and finger tracking (WiLoR) on your GPU, retargets to Mixamo characters, and exports to Blender or Unity.