Trending AI projects

GEO Citation Lab是一个公开实证研究工作台，提供大规模跨平台引用数据、61万条中文引用记录、研究论文集和在线测试，用于研究AI搜索引擎如何选择信源、吸收内容及呈现实体。

PasteGuard is a self‑hosted proxy that redacts personal data and secrets from requests before they reach LLM providers (ChatGPT, Claude, Gemini, etc.). It runs via Docker, offers a dashboard, and supports both masking‑only and routing‑to‑local‑LLM modes.

Intern-Decision is a model that turns a state, optional images, and typed questions (choice, score, yes/no) into calibrated decision probabilities. It fine-tunes Qwen3.5's language backbone, includes training and two inference backends, and ships with a distribution calibration benchmark and a browser demo.

A cloud-native data pipeline orchestrator for managing the development and maintenance of data assets like machine learning models and tables with integrated lineage and observability.

StarVLA is a modular, open-source codebase for developing Vision-Language-Action models for robots. It offers plug-and-play components, multiple VLA frameworks, training recipes, and benchmark integrations, with a focus on rapid prototyping and easy customization.

The Robot Operating System (ROS) is an open-source set of software libraries and tools designed to help developers build robot applications.

AIGCPanel is a cross‑platform desktop app (Electron + Vue 3) that bundles open‑source AI models for digital‑human video, multi‑model TTS/voice cloning, speech‑to‑text, and 25+ audio‑video utilities. It offers a visual drag‑and‑drop workflow editor and live‑stream comment automation (VIP tier), plus a CLI for scripting. Users install models via a built‑in market and can run everything locally or connect to remote/cloud services.

A centralized management tool for AI agent skills that allows users to install once and sync them across dozens of different AI coding tools and devices.

A Codex Plugin providing AI agents with a set of specialized skills and tools to perform automated video editing operations via the chengfeng-videocut runtime.

A cross-platform live speech translation app that enables real-time two-way communication in bilingual meetings using either cloud APIs or fully offline local inference.

An open-source agent workspace for knowledge work and software development that allows users to automate workflows and build AI tools using a choice of models and providers.

Emacs Agent Shell is an Emacs package that provides a native REPL‑style buffer for interacting with any LLM coding assistant that implements the Agent Client Protocol (ACP). It launches the provider’s CLI (Claude, Codex, Gemini, etc.) in ACP mode, handles authentication, environment setup, and even image/screenshot input, letting developers chat with AI agents directly inside Emacs.

An API skill package that enables AI coding tools to interact with EasyEDA Pro for automated debugging and extension development.

nao is an open-source agentic BI builder that allows data teams to create a controlled context layer for AI agents, enabling company-wide, accurate natural language data analysis.

JADX‑AI‑MCP is a Java plugin for the JADX decompiler plus a Python MCP server that lets large language models (e.g., Claude) interact with a decompiled Android APK in real‑time. Users can ask the AI to explain code, find vulnerabilities, rename obfuscated symbols, fetch resources, and even debug, all from within the JADX GUI.

A minimal, self-evolving autonomous agent framework that allows LLMs to control a local computer and autonomously grow a personal skill tree from a small set of atomic tools.

LumiBot is an open‑source Python framework that lets you write a single trading strategy and run it as a backtest, paper‑trading bot, or live bot. It adds a built‑in LLM‑agent runtime (research, debate, risk‑check, execution) while keeping a deterministic Python gate before any broker order. Supports stocks, options, futures, forex, crypto and prediction markets, with integrations for 12 brokers (Alpaca, IBKR, Tradier, etc.). Install via pip, generate starter code with `lumibot init`, backtest on free Yahoo data, then paper‑trade or go live using the same code.

FableCut is a full video editor that runs in the browser and exposes its whole timeline as a single JSON file. AI agents (like Claude) can drive editing through an MCP server or REST API while humans watch live updates, enabling collaborative human–AI video production.

An LLM-first SEO analysis toolkit for AI IDEs that uses 89 scripts and 10 specialist agents to automate technical SEO audits and strategic planning.

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 reasoning framework for AI agents that prevents premature convergence by spawning isolated parallel reasoning processes under different cognitive frames to surface non-obvious solutions.

OpenMed is an Apache‑2.0‑licensed SDK for on‑device clinical‑text extraction and PII de‑identification. It provides multilingual NER models, multiple runtimes (CPU, CUDA, Apple MLX, Android ONNX, browser), and integrates with Python, Swift, Kotlin, and REST services. The library lets you run medical entity detection and safe‑harbor‑aligned de‑identification locally, with optional batch processing and agent‑skill catalogs.

XLA is an open-source machine learning compiler that optimizes models from PyTorch, TensorFlow, and JAX for high-performance execution on GPUs, CPUs, and ML accelerators.

Archon is an open‑source workflow engine that makes AI‑assisted coding deterministic and repeatable. Define development processes (plan, implement, test, review, PR) as YAML workflows; Archon runs them in isolated Git worktrees, mixing AI nodes with deterministic bash steps, and provides a CLI, web UI, and chat‑platform adapters.

Specula is a tool that uses AI coding agents to write TLA+ formal specifications from source code to find and reproduce deep bugs in concurrent and distributed systems.

A self-hosted agent runtime written in Rust that connects LLM providers to multiple communication channels and hardware tools while ensuring user ownership of data and infrastructure.

Coze Studio is an open-source, low-code visual development platform for creating, debugging, and deploying AI agents, workflows, and AI apps.

Gyroflow is an open-source video stabilization app that uses gyroscope (and optionally accelerometer) data — recorded either inside the camera or by an external device — to remove shakiness from footage. It supports many camera brands, offers real-time preview, rolling shutter correction, video editor plugins, and runs on Windows, macOS, Linux, Android, and iOS. The core engine is written in pure Rust, while the UI uses QML; it's licensed under GPLv3 with an App Store exception.

Inno Agent is a personal learning assistant that uses a three-layer memory system and a proactive scheduler to provide personalized, evidence-driven educational support.

A coordination framework that manages AI agents in isolated git worktrees and tmux sessions, using a durable handoff protocol to exchange work.

A Codex skill that researches public demand signals to find and qualify potential first customers for a product, providing verified contact routes and a feedback-driven search process.

An interactive 12-chapter textbook that teaches you to build a LLaMA 3-style GPT from absolute scratch. Every line of code is annotated with WHAT and WHY, and it requires only basic Python — no math or ML background needed. Ideal for developers who want to truly understand how modern LLMs work inside.

Awesome Prompt Engineering is a hand-curated directory of resources for prompt engineering and context engineering with large language models. It gathers research papers, tools, APIs, models, benchmarks, courses, and communities in one place, with a beginner path and sections covering reasoning, security, multimodal generation, and agentic systems.

A curated, quality-gated list of books, courses, papers, frameworks, and tools for developers building generative AI and agent-based systems — plus a section on AI coding agents and multi-agent orchestration.

Agent Vault is an open‑source HTTP proxy that stores secrets and injects them into outbound requests for AI agents, preventing the agents from ever seeing real credentials. It offers a management UI, CLI, per‑agent tokens, pluggable secret back‑ends (including Infisical), egress filtering, request logging, and optional PostgreSQL scaling. Deployable as a single binary or Docker image, it integrates via standard `HTTPS_PROXY` environment variables and is aimed at securing remote‑coding, all‑purpose, custom, and sandboxed AI agents.

The project is a curated list of open-source AI tools, models, and resources, organized into 14 categories. It prioritizes usefulness over popularity, includes projects regardless of star count, and invites contributions.

Petdex is a community-driven platform and desktop companion system that lets you install and display animated mascots that react to your coding agent's activity in real time.

A unified agent sandbox environment that combines a browser, shell, and IDE tools in one Docker container with a shared filesystem for AI agents.

The Simulink Agentic Toolkit is a MATLAB add‑on that connects LLM coding agents (Claude Code, Copilot, Codex, Gemini, etc.) to Simulink via a Model‑Context‑Protocol (MCP) server. It provides a catalog of Model‑Based Design skills and concrete tools (read, edit, test, check, query) so agents can safely manipulate Simulink models, generate tests, and enforce best practices. Installation is automated through a toolbox installer, and the toolkit works with MATLAB R2023a+ and optional MathWorks products (e.g., Simulink Test).

An interactive, version‑checked tutorial repo that teaches users how to use Anthropic’s Claude Code AI coding assistant through step‑by‑step lessons, automated checks, and three skill levels—from basic commit assistance to autonomous multi‑agent CI pipelines.

Steel is an open-source browser API that provides the infrastructure needed to build AI agents and web automation tools, handling browser sessions, proxy management, and page interaction.

A comprehensive collection of Chinese-language study notes for Andrew Ng's Deep Learning Specialization, covering neural network fundamentals, optimization, and advanced architectures like CNNs and RNNs.

An agentic CLI for the Cloudflare API that provides TypeScript-based configuration and JSON interfaces optimized for both human developers and AI agents.

The repository hosts Feynman, a CLI‑based AI research assistant that can perform literature reviews, paper audits, replication planning, and draft writing by orchestrating multiple LLM‑driven agents. It installs via script or npm, connects to model providers, queries many scholarly APIs, and stores results with provenance. Telemetry is optional.

Pixal3D is a high-fidelity 3D asset generator that creates detailed geometry and PBR textures from a single image or multiple views using pixel-aligned back-projection.

vLLM Metal is a plugin that enables the vLLM LLM serving system to run on Apple Silicon Macs using the MLX/Metal backend. It provides Metal‑optimized attention kernels (paged varlen, M5 NAX prefill) and supports large models such as Qwen‑3.8‑27B. Install via Homebrew or a curl script, then use the regular `vllm` command on macOS 15+.

Hatchet is an open‑source orchestration engine that lets you run reliable background tasks, AI‑agent pipelines and durable workflows. It stores all state in PostgreSQL, provides retries, scheduling, DAGs, rate‑limiting, priority queues and a real‑time UI, and offers SDKs for Python, TypeScript, Go and Ruby. You can use it as a managed cloud service or self‑host it via Docker.

A plugin for AI agents that provides a library of 48 specialized marketing, SEO, and advertising workflows with live data integrations via MCP.