Trending AI projects

kicad‑happy is a Python‑only, AI‑enhanced KiCad design‑review tool. It parses schematics/PCBs, builds a power‑tree, runs rule‑based checks (ESD, decoupling, trace widths), generates SPICE test‑benches, evaluates EMC compliance, cross‑references datasheets, and creates a full BOM with supplier queries. It can be used locally, as a GitHub Action, or as plug‑ins for Claude Code, OpenAI Codex, Google Antigravity, and opencode, enabling natural‑language queries like “Check my boost converter for EMI”.

A professional financial skill set for Claude adapted for the Chinese A-share market, integrating Wind, iFind, and AkShare for automated financial analysis and reporting.

A multi-agent offensive-security framework that turns existing AI coding agents into automated zero-day hunters for authorized security testing.

editonny is an open‑source Claude Code / Codex skill that lets you edit videos by describing the desired changes in plain language. It runs locally, handling pause‑cutting, caption generation, audio fixing, re‑framing, and inserting pre‑built motion‑scene animations. Installation is a one‑request paste into the agent (or a manual terminal install). After confirming the proposed edit plan, the skill produces a preview and then exports a polished MP4, all without needing external LLM APIs.

An AI creative workspace for images and videos that combines an AI assistant, a visual canvas for organization, and integration with various generative models.

A client-side JavaScript library that embeds an AI agent into web pages for natural language DOM manipulation and automation without requiring headless browsers.

Happier is a multi-device control room that orchestrates teams of AI coding agents, allowing you to run, monitor, and steer development sessions from your desktop, phone, or browser.

A self-hosted AI workspace that transforms chatbots into digital teams by providing a governed runtime for humans, AI actors, and remote agents to collaborate in shared conversations.

An autonomous agent framework that builds, generates assets for, and tests games across multiple engines like Godot, Bevy, and Babylon.js based on user prompts.

ContextForge is an open-source registry and proxy that federates MCP, REST, and gRPC services into a single, governed endpoint to simplify tool and API management for AI agents.

An open-source relational database that provides MySQL compatibility and advanced features like native vector search and pluggable storage engines.

A public Model Context Protocol (MCP) server that lets AI coding assistants (Cursor, Claude, Codex, etc.) browse Solana’s API, query on‑chain data, and build unsigned transaction payloads via a single HTTP endpoint. The repo contains the server’s registry metadata, plugin manifests for various AI platforms, example configuration snippets, and a GitHub Action‑based publishing workflow.

Agent Zero is an open‑source Docker‑based framework that gives an LLM‑driven AI agent a full Linux desktop, a browser with DOM‑annotation, live document editors, and a plugin ecosystem. Agents can run GUI apps (e.g., Blender, LibreOffice), edit files, browse and annotate web pages, and even spawn sub‑agents, all while you watch and intervene through a web UI. Installation is via a cross‑platform launcher or a one‑line script, and the system supports project‑scoped workspaces, host‑machine bridging, time‑travel snapshots, and extensive community plugins.

A methodology and set of patterns for structuring AI agent workflows using a filesystem hierarchy instead of complex orchestration frameworks, enabling human-in-the-loop review and high observability.

Apache Ossie defines a JSON/YAML‑based, vendor‑neutral semantic model spec for AI, analytics, and BI tools, enabling consistent KPI definitions across platforms.

A library of 34 reusable HTML slide templates specifically designed for AI coding agents to automatically generate beautiful presentation decks.

Quicksand is an asynchronous Python library that provides secure, isolated QEMU virtual machine sandboxes for running and testing AI agents without requiring root or Docker.

A browser automation plugin for DeepSeek Harness that provides a real-time visual observation and manual control interface for AI agents operating within a Chromium browser.

ClawGod is a runtime patcher that modifies the official Claude Code binary to unlock hidden commands, remove safety/usage restrictions, add multi‑agent features, and fix reliability issues. It works by extracting Claude’s embedded JavaScript, applying regex patches, and running the patched code under a regular Bun runtime, auto‑re‑patching on updates.

TrueForge is an open‑source Node.js harness that runs LLM agents, handling model calls, tool integration, sandboxed execution, session storage, approvals, and UI. It works with any OpenAI‑compatible model, offers a chat UI, HTTP API, and TypeScript SDK, and can run locally (SQLite) or in hosted mode (Postgres + Redis).

Cezar is an open‑source, self‑hosted dashboard + CLI that orchestrates multiple AI coding agents (Claude Code, Codex, Copilot, etc.) to run parallel, autonomous coding tasks. It uses Git worktrees for isolation, YAML workflows for step‑by‑step pipelines, and streams live output (prompts, tool calls, token usage, cost) to a responsive UI that works on desktop and mobile. Features include autonomous runs, variant comparison, GitHub/Jira/Linear integration, scheduling/automation, and zero‑database storage in plain files.

Desert Ant Core is a multi‑platform SDK (Swift, Kotlin, JavaScript/TypeScript) that provides on‑device AI models for speech enhancement, transcription, language ID, emoji suggestion, PII redaction, NSFW image detection, and more. Models are downloaded from Hugging Face, cached locally, and can run offline. The library supports iOS/macOS, Android, Windows, browsers (WebAssembly), and Node, with a CLI for terminal use. Free for most apps under a source‑available license; commercial scaling needs a paid license.

A comprehensive library of 89+ AI prompts and templates designed to help graduate students and researchers streamline their workflow from topic selection to manuscript writing.

Atomic Chat is a local AI application and inference engine that allows you to run open-weight models privately while providing an OpenAI-compatible API for your favorite developer tools and agents.

An open-source text-to-CAD web app that uses AI to transform natural language and images into parametric 3D models via OpenSCAD.

A Python project that runs an AI digital-human live stream: it filters and queues viewer chat, routes messages to actions (reply, outfit change, gesture), and drives the Vidu S2 real-time avatar API, with a local workbench, OBS integration, and direct streaming support.

Aletheore is a deterministic, evidence‑grounded code‑intelligence toolkit. It scans a repository to produce a structured evidence file (dependency graph, secrets, git history, API map, etc.), which can be queried locally, fed to LLMs for audited reports, or exposed via a lightweight MCP server for AI agents. The CLI is free and offline; a paid GitHub App adds automated PR reviews, AI‑generated docs, and monitoring.

Tulpa is a local bridge that connects QQ and WeChat data to AI agents, enabling them to search chat history, summarize discussions, and participate in group chats with custom personas.

An MCP server that enables AI assistants to query Devnors Data APIs for legal, enterprise, research, and content data.

SolarRenderer converts rough structure videos (grey boxes, semantic maps, skeletons, depth, meshes, even unseen sketches) into finished videos in four generation steps, with optional first-frame and prompt control over style. It runs on MiniMax-H3 with two LoRAs, supports block-by-block streaming, and offers an online demo, a reference script, ComfyUI workflows, and a Reactor streaming integration. Code is Apache 2.0; weights follow MiniMax H3's community license with territorial restrictions.

Reachy Mini is an open‑source robot kit (wireless or USB) with Python and JavaScript SDKs, simulation support, and a Hugging Face app store for AI‑driven behaviours. It lets makers and AI developers quickly program expressive motions, integrate LLMs, and share web‑based robot apps.

A benchmark for evaluating large language models on their ability to resolve real-world GitHub software issues by generating functional code patches.

A local AI server stack that provides a one-command installation of inference engines, chat interfaces, and automation tools for private, self-hosted AI.

FlashMLA is a high-performance library of optimized attention kernels for DeepSeek models, providing accelerated sparse attention, quantized KV cache support, and fused operations for NVIDIA and Huawei hardware.

Argent is a CLI that lets AI assistants control iOS simulators, Android emulators/devices, TV platforms, and Electron/Chromium apps. It supports UI interaction, end‑to‑end test recording, visual‑regression diffing, profiling, and debugging, with built‑in React Native support.

A virtual terminal environment for AI agents that maps external services and data sources into a unified, secure, and POSIX-compliant filesystem.

An open source AI toolkit for building real-time, on-device voice agents and applications with low-latency speech-to-text and text-to-speech capabilities.

A curated, categorized directory of AI-powered coding tools, ranging from intelligent code editors and autonomous agents to productivity and DevOps utilities for software engineers.

A comprehensive prompt-engineering library for Higgsfield AI that helps users generate high-quality, cinematic video and images through structured templates and professional production workflows.

An open-source autopilot stack for drones and unmanned vehicles that provides a modular flight control system supporting various airframes and hardware.

A production-ready orchestration layer for running high-performance, cross-platform LLMs on-device using hardware acceleration.

Rungic turns an Android phone into a Linux desktop computer and adds an AI agent that can autonomously operate desktop applications to complete tasks for the user.

FastGPT is an open‑source platform for building AI agents and RAG‑enabled applications. It offers a visual workflow editor, built‑in knowledge‑base handling, plugin hot‑reload, debugging tools, and can be launched instantly with Docker or via a hosted cloud service.

Oh Story DSH is a DeepSeek Harness plugin that adds four AI‑driven creative workbenches—novel writing, short‑drama scripting, novel‑to‑game adaptation, and video recap—using bundled Skills and Roles. It runs inside the DSH web UI, works on plain files, and optionally calls external media APIs. Install with a single `npx` command (Node 24+, pnpm) and configure model/API keys as needed. MIT‑licensed.

A desktop workspace that integrates AI agents directly into Markdown documents, structured tables, and whiteboards, allowing for reviewable and automated content workflows.

An AI agent for hardware development that designs, documents, and validates KiCad PCBs from natural-language prompts while ensuring ERC/DRC compliance.

OpenChatX is a local agent platform that connects to ChatGPT via official MCP tunnels, enabling advanced capabilities like file management, shell execution, and persistent cross-session workflows.

FinRobot is an open‑source, multi‑agent AI platform for high‑stakes financial analysis. It separates deterministic Python‑based valuation (DCF, LBO, etc.) from LLM‑generated narration, orchestrates specialized agents (data, analysis, modeling, synthesis, reporting, plus debate agents), and ships three versions: a production desktop app (V2), a self‑hosted web report generator (V1), and an educational AutoGen library (V0). The system integrates multiple market data sources, provides 56 analyst playbooks, and can be run locally via a CLI, a browser UI, or a native Tauri desktop client.