A set of modular AI agent skills that implement a professional engineering workflow, moving project state from chat sessions into repository files for persistence and team collaboration.

A simulated bash environment with a virtual filesystem that allows for safe execution of bash-like commands in a sandbox.

A desktop application that uses the RMBG-1.4 model to provide local AI-powered background removal, batch processing, and image editing tools.

An AI-engineering foundation framework that provides an architecture-aware harness and spec-first development to build consistent, professional business applications.

An extension of veRL for training LLM agents via reinforcement learning, featuring a step-independent rollout mechanism to efficiently handle long-horizon, multi-turn tasks.

A browser automation tool that lets AI agents control your existing Chrome browser session, preserving logins and extensions to bypass bot detection.

An MCP server that connects AI assistants to the Hevy workout tracking app, allowing users to analyze training progress and manage workouts via natural language.

A Go-native framework for building production AI agent systems with support for graph workflows, multi-agent collaboration, persistent memory, and self-evolving skills.

An open-source CLI coding assistant from Mistral AI that provides a conversational interface to explore and modify codebases using natural language and integrated tools.

A modular workstation for 3D Gaussian Splatting that integrates training, real-time visualization, editing, and export in a single native application.

A Java library for building stateful, multi-agent AI applications using cyclical graphs, designed to work with LangChain4j and Spring AI.

A cross-platform Golang library for desktop automation and AI computer use, enabling programmatic control of the mouse, keyboard, and screen.

An event streaming platform that replaces complex data stacks with a single system for real-time data ingestion, incremental processing, and low-latency serving for AI agents.

Claudexor is a local‑first Node.js control plane that unifies multiple AI‑coding services (Codex, Claude Code, Cursor, etc.) behind a single typed interface. It provides read‑only queries, patch‑generating write turns, best‑of‑N races with cross‑model review, credential profiles with live quota tracking, deterministic gates for protected files, and a macOS desktop app. All state lives on your machine, there is no telemetry, and the system can run remotely over SSH. The project is actively maintained (v3.10.2) and intended for developers or agents who need auditable, cost‑controlled AI‑assisted coding.

An open-source, JavaScript-driven desktop robot platform based on M5Stack that allows users to create and customize robot behaviors via browser-based tools and MODs.

A self-hosted, open-source alternative to LangSmith Deployments that allows you to run LangGraph agents on your own infrastructure using the same SDK and APIs.

A family of conditional generative models that repurpose pretrained diffusion models like Stable Diffusion for high-resolution monocular depth, surface normal, and intrinsic image analysis.

A desktop AI writing system for long-form fiction that uses structured memory and local semantic search to maintain narrative consistency across complex stories.

PyGPT is an open‑source desktop AI assistant (Linux, Windows, macOS) that lets users chat with a wide range of LLMs—both cloud‑based (OpenAI, Gemini, Claude, etc.) and locally hosted via Ollama. It supports multimodal interactions (text, files, images, video, speech), RAG through LlamaIndex, multi‑agent orchestration, code execution, and extensible plugins, all wrapped in a Qt‑based GUI.

An MCP server that connects Figma design data to AI coding agents, enabling them to generate accurate code and modify designs based on design system context.

Rulesync is a Node.js CLI that centralises AI‑assistant configuration. You write unified `.rulesync/` rule files and the tool can generate, import, export, or convert native config files for dozens of AI coding tools (Claude, Copilot, Cursor, etc.). Installation is via npm, Homebrew, or a single‑binary script. Features include selective generation, broad tool support, deprecation handling, and a programmatic API. MIT‑licensed.

ExecuTorch is an open‑source PyTorch‑based runtime for deploying AI models on edge devices—from phones to micro‑controllers. It exports models directly from PyTorch, compiles them ahead‑of‑time into a tiny `.pte` file, and runs them on a variety of hardware back‑ends via lightweight C++/Python/Swift/Kotlin APIs. It supports vision, speech, and LLMs (e.g., Llama) and is used in Meta’s production apps.

A Nix flake-based reverse engineering environment that integrates with Claude Code to provide automated, context-aware toolchains for analyzing binaries and network captures.

LinguaGacha is a cross‑platform desktop app that uses large‑language‑model APIs (OpenAI, DeepSeek, Anthropic, Google, or local models) to translate subtitles, e‑books, and game scripts in dozens of languages with a single click. It offers an “AGENT” workflow for automatic glossary extraction, batch translation, and auto‑review, preserving formatting and code‑style elements.

A plug‑in marketplace for LLM‑based coding assistants (Claude Code, Gemini, Antigravity, etc.) that supplies token‑efficient, context‑engineering patterns—reflection loops, sub‑agent judges, specification‑first development, code‑review agents, and Git helpers—to make AI‑generated code more reliable and easier to integrate into CI pipelines.

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

A minimalist ML framework for Rust that enables lightweight, Python-free serverless inference and browser-based model execution.

Google's high-performance on-device runtime for deploying ML and Generative AI models across mobile, web, and IoT platforms with advanced hardware acceleration.

TrustGraph is an open‑source platform that converts enterprise data into an RDF/OWL‑based hypergraph, letting LLM agents query structured, provenance‑rich context instead of relying on fuzzy vector similarity. It offers data ingestion pipelines, configurable “hyperflows” for multi‑step agent work, fine‑grained workspace/collection isolation, an open LLM inference stack, and flexible deployment (Docker, Kubernetes, or SaaS). The project includes a web UI, TypeScript client libraries, and is Apache‑2.0 licensed.

A foundation GUI agent that unifies mobile, computer, and web navigation into a single model, enabling autonomous cross-platform task execution and proactive services.

A reverse-engineered asynchronous Python wrapper for the Google Gemini web app that enables programmatic access to features like image/video generation and Gemini Gems.

An MCP-based server that allows Large Language Models to search for 12306 Chinese railway ticket information, including transfers and stop-overs.

A toolkit for automatically discovering and improving AI skills and prompts for coding agents, evolving general agents into specialists using benchmark-driven feedback.

An Android translator app that performs text, PDF/ODT document, and image translation completely offline using on-device models.

A native macOS console for herdr that allows users to manage and interact with AI coding agents across local and remote machines via a full TUI interface.

An open-source, self-hosted platform for tracking and optimizing brand visibility and citations across AI answer engines like ChatGPT, Perplexity, and Gemini.

An open-source toolkit for Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO) that audits and improves a website's visibility and citability by AI answer engines.

LLPlayer is a Windows‑only C# media player that adds AI‑driven subtitle generation (Whisper ASR), real‑time translation (via cloud or local LLMs), OCR for bitmap subtitles, dual‑subtitle display, word lookup, and online video support. It targets language learners and is open‑source under GPL‑3.0.

A scikit-learn compatible tabular foundation model that enables zero-shot classification and regression on mixed-type tabular datasets using in-context learning.

An AI Agent skill that automates deep research across eight major Chinese internet platforms, aggregating and analyzing content from the last 30 days to generate verifiable reports.

Embabel Agent Framework is a Kotlin/Java library for building goal‑oriented, LLM‑augmented agents on the JVM. It provides strong‑typed actions, dynamic planning (GOAP/Utility AI), Spring integration, and three execution modes (focused, closed, open). Developers write agents via annotations or a DSL, and the framework automatically plans and replans actions, mixing LLM calls with regular code. The project is fully open‑source (Apache 2.0), Maven‑published, and supported by docs, CI, and a Discord community.

A single-file, offline-first personal website that transforms AI APIs into immersive emotional companions through a suite of interactive modules and local data storage.

An AI-friendly macOS terminal emulator based on WezTerm that provides pre-configured shell ergonomics and integrated AI assistance for command generation and error fixing.

An AI-powered pre-publication compliance checker for Douyin, Xiaohongshu, and WeChat Channels that helps creators avoid platform violations by analyzing context and official rules.

deja‑vu is a local‑only memory layer for code‑generation agents (Claude Code, Codex, Cursor, Copilot, etc.). It indexes the transcript files those agents already write, builds a fast searchable index, and automatically injects relevant past decisions into any agent session. Features include retroactive natural‑language search, cross‑agent recall, compaction‑aware persistence, state tracking, staleness warnings, sync/handoff between machines, and automatic redaction of secrets. Install via a one‑line script, Homebrew, Scoop, Go, or NPM; `deja install --auto` wires a tiny MCP server into detected agents. Use commands like `deja "jwt refresh token"`, `deja wip`, `deja blame <file>`, `deja sync ssh <host>`, and optional semantic embedding via a local LLM. All processing stays on the machine, with built‑in privacy safeguards.

An engineering-driven writing environment for long-form fiction that uses a world-state engine and plot tracking to prevent AI hallucinations and narrative inconsistency.

DeterminFlow is an open‑source, AGPL‑licensed framework for building, running, and monitoring production‑grade AI workflows. It breaks complex pipelines into isolated LLM‑Agent nodes, scripts, API calls and human approvals, providing versioning, checkpoint‑based recovery, per‑node token accounting, and a visual editor. The tool is aimed at content generation, data‑centric AI pipelines, and enterprise automation, and can be installed via desktop binaries, run from source, or deployed with Docker.

A collection of custom nodes and interface improvements for ComfyUI that simplifies complex workflows through better organization, advanced switching, and condensed loaders.