A framework for local-first personal AI agents that run on personal devices by default, reducing reliance on cloud APIs while optimizing for energy and latency.

A library of 818 structured cybersecurity skills and playbooks designed to give AI agents the domain knowledge and decision-making workflows of a senior security analyst.

An open-source, privacy-focused alternative to Notebook LM that lets you organize research and chat with your data using local or cloud AI models.

A settings plugin for DeepSeek Harness that removes restrictive system prompts and refusal patterns to enable jailbreaking and custom personas.

Cockpit Tools is a cross‑platform desktop manager that lets you store, switch, monitor and run multiple accounts of many AI‑coding assistants (Antigravity IDE, Codex, GitHub Copilot, Windsurf, Kiro, Cursor, Grok CLI, CodeBuddy, Qoder, Trae, Zed, ZCode, etc.). It provides a unified dashboard with real‑time quota tracking, one‑click account switching, multi‑instance isolation, batch operations and optional local WebSocket integration. Install via native installers or Homebrew; data stays on your machine and the project is released under a non‑commercial CC‑BY‑NC‑SA 4.0 license.

An AI-powered workbench for short drama production that automates the workflow from scriptwriting and storyboarding to final video generation.

A VS Code extension and GitHub Copilot app canvas that analyzes local AI coding session logs to provide insights into prompting patterns, context health, and developer productivity.

A vetted community marketplace for plugins that extend the capabilities of Claude Cowork and Claude Code.

A comprehensive, automatically updating library of AI Agent skills aggregating official and community-curated prompt sets for diverse professional domains.

An open-source background coding agent system that executes software engineering tasks in isolated sandboxes and submits results as Pull Requests.

An ultra-lightweight, self-hosted personal AI agent framework that combines long-term memory, MCP integrations, and scheduled automation across WebUI, terminal, and various chat apps.

Whisper is OpenAI’s open‑source, transformer‑based speech‑to‑text system that handles transcription, language identification, and translation across many languages. It ships as a pip‑installable package with six model sizes (tiny → large, plus a fast “turbo” variant). The CLI (`whisper …`) and a simple Python API (`whisper.load_model(...).transcribe(...)`) let you turn audio files into text with just a few commands. The project runs on Python 3.8‑3.11, PyTorch, and requires ffmpeg (and optionally Rust for the tokenizer). All code and model weights are MIT‑licensed.

An open-source coding-agent CLI that provides a unified terminal workflow for cloud and local LLM providers, supporting tool-driven coding and codebase intelligence.

An open-source SEO analysis plugin for Claude Code that uses parallel specialist agents to perform technical, content, and AI-search optimization audits.

jcode is a terminal‑based AI‑agent harness that emphasizes ultra‑low RAM usage and sub‑50 ms startup. It embeds each conversation turn into a vector graph for automatic memory recall, offers side‑panel file/diagram views, and supports a “swarm” mode where multiple agents collaborate on the same codebase. The tool works with many cloud providers (Claude, OpenAI, Gemini, Copilot, etc.) and any OpenAI‑compatible endpoint, and can be installed via a one‑line script on macOS, Linux, or Windows.

A desktop and code-based platform for running and fine-tuning AI models locally, offering significantly faster training speeds and lower VRAM requirements.

A template that enables AI coding agents to clone any website into a clean Next.js application using a multi-phase reconnaissance and build pipeline.

Langflow is a visual platform for building and deploying AI agents and workflows, allowing developers to design, test, and export AI logic as APIs or MCP servers.

CosmoEdge 1.1 is an Apache‑2.0 C++ edge‑AI engine for video analytics and vision‑language models. It runs on Sophon (BM1688/CV186X), Rockchip (RK3576/RV1126B) NPUs and on x86/macOS via Docker, offering a browser‑based pipeline builder, REST/MQTT/WebSocket integration, and a “Model Guard” for protected commercial models. The repo includes Docker scripts for building device‑specific runtimes, benchmark reports, and extensive docs.

An open-source, local-first chat application that lets you manage a team of AI agents as contacts, integrating with local CLIs and providing controlled computer access.

Switchyard is a pre‑1.0 NVIDIA‑NeMo project that routes each LLM request to the cheapest model that can still solve the task. It offers a Rust library, a Python package, a NeMo Relay plugin, and a demo OpenAI‑compatible proxy. Configurable routing algorithms (stage, capability, escalation, etc.) let you automatically fall back from cheap to strong models, cutting token cost by up to ~30 % while keeping accuracy near that of a single high‑end model. The code is Apache‑2.0 licensed and intended for integration into existing LLM gateways or custom harnesses.

Cua is an open‑source stack that gives AI agents the ability to control desktop applications in the background across macOS, Windows, Linux and Android. It includes cross‑OS drivers, a Python sandbox SDK for spinning up VMs/containers, a benchmark suite (Cua‑Bench) for evaluating computer‑use agents, and Lume for fast macOS virtualization on Apple Silicon. The project is MIT‑licensed and targets researchers and developers building autonomous agents that need to interact with real GUIs.

A macOS application that enables Apple Silicon Mac users to run the Omarchy desktop environment as a native, hardware-accelerated ARM64 virtual machine.

A high-performance, secure sandbox service for AI agents that provides hardware-isolated MicroVMs with sub-60ms startup times and E2B SDK compatibility.

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.

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

A self-hosted, OpenAI-compatible LLM router that optimizes costs and reliability using automatic model selection and a managed fallback safety net.

A game mod that uses NVIDIA AI to enhance lighting, detail, and color in supported games, allowing for customizable neural rendering effects.

A tool that enables NVIDIA DLSS 5 Neural Rendering on AMD Radeon RX 9000 and 7000 series GPUs for DirectX 12 games using FSR.

A document parsing and retrieval system that converts unstructured files into a hierarchy-native memory substrate for AI agents to navigate and cite.

A control-plane platform for deploying, routing, and securing A2A-speaking AI agents with built-in observability and secret management.

An open-source AI image creation workbench that integrates an infinite canvas, multi-modal AI generation, and agentic canvas manipulation into a single workspace.

A highly optimized inference server for Qwen3.8-Flash-Next on AMD Strix Halo hardware, delivering significantly faster prefill and decode speeds via hardware-specific kernels.

A design skill for AI coding assistants that uses a rigorous anti-pattern rule-set and 21 themes to create professional UI that avoids the generic AI-generated look.

An all-in-one Agent chatbot platform that integrates LLMs and AI agents into mainstream instant messaging apps like Telegram, Slack, and Discord.

An open-source AI Development Environment (AIDE) that lets developers visually build, debug, and deploy high-performance AI pipelines using a C++ runtime.

An AI-powered cybersecurity automation platform that integrates 150+ security tools into an MCP framework for autonomous penetration testing.

A TypeScript library for WebGPU that provides typed shader imports and a unified API for running GPU code in the browser, headless Node, and test environments.

SkillOpt is a framework that treats agent skill documents as trainable parameters, using an optimizer model to iteratively improve them without changing model weights.

A production-grade, Rust-native trading engine that provides a unified event-driven architecture for both research and live multi-asset trading.

A coding-agent skill that turns an AI agent into a security auditor using a structured six-phase process of reconnaissance, hunting, and adversarial validation.

OpenSenseNova’s SenseNova‑U1 series are open‑source, native‑unified multimodal models (text + image) built on the NEO‑unify architecture. They support high‑quality 4K image generation, dense‑text infographic creation, image editing, VQA, and interleaved text‑image tutorials. The repo provides training code, pre‑trained 8 B‑parameter checkpoints (including community‑quantized GGUF files), example inference scripts, and deployment guides, all under an Apache 2.0 license.

A set of opinionated instructions and a workflow contract for AI coding agents to ensure high-quality, predictable code and professional communication.

A local-first AI e-book reader that uses RAG and semantic search to help users chat with their library and build a private knowledge base.

An open-source platform for running persistent AI teammates with their own memory and the ability to operate browsers, terminals, and desktops.

A framework for distilling methodologies from books, videos, and podcasts into structured, executable AI skills that agents can invoke for real-world problem solving.

A coding-agent-driven framework that turns prompts into editable, structured visual designs using HTML and semantic layers instead of flattened images.

A high-performance file and content search toolkit that provides typo-resistant, frecency-ranked search for humans and AI agents.