A drop-in prompt compression library that reduces LLM input-token costs by 10-40% using loss-tolerance routing to protect critical code and tool calls while compressing prose.

An open-source, cross-platform AI client that allows users to self-host their AI interface and connect to local or frontier model providers to avoid vendor lock-in.

A packaged desktop distribution of AstrBot that bundles the WebUI and backend runtime for fast local installation and use of ChatUI and plugins.

A comprehensive collection of foundation models and architectures from Microsoft focusing on large-scale self-supervised pre-training across text, vision, speech, and multimodal tasks.

Apify Agent Skills is a set of plug‑in style “skills” that let LLM agents (Claude, Cursor, Gemini, etc.) invoke Apify’s web‑scraping/automation platform. Five core skills cover universal scraping via 130+ curated Actors, creating new Actors, turning existing code into Actors, auto‑generating output schemas, and building Apify integrations. Install with a single `npx skills add …` command, then prompt the agent to scrape sites, generate leads, do market research, etc. Requires an Apify account, the Apify CLI, and Node 20+. Licensed Apache‑2.0.

Moss is a sub-10 ms semantic search runtime that embeds retrieval and embeddings directly into an application's process to eliminate network latency for real-time AI agents.

Text Grab is a Windows‑only desktop app that captures any visible text (screenshots, PDFs, UI elements) using local OCR (WinAI, WinRT OCR, or Tesseract) and provides built‑in cleanup, spreadsheet editing, regex‑based extraction, reusable grab templates, bulk folder processing, and a Chrome/Edge extension. All processing stays on‑device, with optional NPU‑accelerated inference on Copilot+ PCs. Install via Microsoft Store, GitHub releases, or package managers; source can be built with Visual Studio or the .NET SDK.

A Python package for aircraft and engineered system design and optimization using automatic differentiation and differentiable physics models.

Bol na is an open‑source framework that orchestrates speech‑to‑text, LLM, and text‑to‑speech services to build voice‑first assistants that can place and receive phone calls. It ships Docker‑compose containers for the core server, telephony webhook (Twilio/Plivo), Redis, and ngrok, and provides a Python SDK (`Assistant`) for building streaming pipelines. The platform is provider‑agnostic (supports Deepgram, Azure, OpenAI, ElevenLabs, AWS Polly, etc.) and is extensible to new telephony or audio services.

A web-based viewer for Gaussian Splatting scenes that allows developers to embed high-performance 3D splat assets into static websites using WebGPU or WebGL.

A collection of portable knowledge packs that teach AI coding agents Home Assistant best practices to prevent common configuration errors.

DoWhy is a Python library for end‑to‑end causal inference. It lets you define a causal graph, automatically identifies the causal effect, estimates it with a variety of statistical methods, and runs refutation tests to check robustness. The library also supports graphical causal models for root‑cause analysis, counterfactuals, and interventional sampling. Install via `pip install dowhy` and follow the four‑step workflow (model → identify → estimate → refute).

A professional SEO Agent Skill for AI assistants that performs technical audits, traffic diagnostics, and visibility optimization for traditional search engines and AI search platforms.

A scalable, robust model-based reinforcement learning algorithm for continuous control that can handle multiple tasks across different domains and embodiments using a single agent.

Banana Claude is a Claude Code plugin that lets you write a natural‑language request, see a detailed plan (prompt, model, cost estimate) and, after an explicit approval, call Google’s Gemini image models to generate or edit visual assets. It emphasizes offline‑first planning, privacy‑safe API key handling, and one‑time approvals to keep costs transparent.

MCP Gateway is an open‑source Kubernetes‑native reverse proxy and management layer for Model Context Protocol (MCP) servers. It provides a REST control‑plane for creating, updating, and monitoring MCP adapters and tools, session‑aware routing, Azure Entra ID RBAC, a built‑in React portal, and optional LLM‑driven agents. Deploy locally with Docker/K8s or with a one‑click Azure template.

An open-source, local-first AI journal that uses a multi-agent system to automatically organize life fragments into a structured, private knowledge base.

An AI-powered toolkit for e-commerce sellers to automatically analyze product photos and generate a complete set of professional marketing images across various styles and platforms.

A comprehensive collection of examples and tutorials for deploying and fine-tuning Liquid AI's open-weight Liquid Foundation Models (LFMs) on mobile, desktop, and browser environments.

Pentest‑AI is an open‑source Python tool that lets you run LLM‑driven penetration‑testing against web applications. It generates probes via a connected AI model, then uses a deterministic oracle to reproduce each exploit and verify it against a control request. Verified findings are stored in replayable proof capsules and can be exported as SARIF for CI/CD pipelines. The package works with cloud models (Codex, Claude Code) or local Ollama models, is installed via `pipx`, and is MIT‑licensed.

Self‑hosted web app that photographs your clothes, uses a multimodal LLM (Ollama, OpenAI, etc.) to tag items and generate daily outfit suggestions based on weather and occasion. Docker‑compose (or K8s) deployment, PostgreSQL + Redis backend, Next.js frontend, fully internationalised and privacy‑first.

An autonomous financial research agent that uses task planning and real-time market data to perform deep financial analysis.

A framework for computing embeddings, retrieval, and reranking using state-of-the-art dense, sparse, and multi-vector encoder models.

An agent-ergonomic browser automation CLI that wraps chrome-devtools-mcp to provide token-efficient, high-success web navigation and interaction for AI agents.

An autonomous agentic AI system that monitors Git commits for vulnerabilities and manages Jira tickets to ensure continuous compliance with the EU Cyber Resilience Act.

A demo repository for running Bonsai and Ternary-Bonsai language models locally, featuring extreme 1-bit and 2-bit quantization for high-performance inference on consumer hardware.

An agentic LLM designed for autonomous data science, capable of automating the entire pipeline from data preparation to the generation of professional research reports.

A proxy server that exposes an Anthropic-compatible API backed by Antigravity's Cloud Code, enabling the use of Claude and Gemini models with the Claude Code CLI.

A self-hosted server that uses Whisper and LLMs to automatically detect and remove advertisements from podcast episodes, serving them via clean, ad-free RSS feeds.

A Three.js agent skills pack that provides AI agents with a library of sophisticated graphics implementations to produce professional-grade 3D visuals.

A 15TB collection of diverse numerical physics simulations and a benchmarking suite designed to train and evaluate machine learning surrogate models for computational sciences.

An AI-powered, cross-platform desktop SQL client that supports multiple databases and helps users write and optimize queries.

BiliSum is a video-to-knowledge tool that converts Bilibili, YouTube, and local videos into transcriptions, visual notes, and a local RAG-powered knowledge base.

A command-line tool for LM Studio that allows users to manage local AI models and control the API server via the terminal.

A framework for reinforcement learning and search/planning in games, providing a collection of environments and algorithms for multi-agent and single-agent research.

LLM Compressor is a Python library that quantizes and prunes large language models into the `compressed‑tensors` format, enabling memory‑efficient deployment with vLLM. It supports many low‑precision formats (NVFP4, FP8, INT4, etc.), several PTQ/GPTQ algorithms, DDP and disk‑offloading for huge models, and ships pre‑quantized checkpoints for popular LLMs.

A security scanner for AI Agent Skills that detects prompt injection, data exfiltration, and malicious code using static analysis, behavioral dataflow, and LLM-based semantic analysis.

CodeWiki is an open‑source Python framework that uses LLMs (OpenAI, Anthropic, Azure, Bedrock, Atlas Cloud, or Claude/Codex subscription) to automatically generate comprehensive, architecture‑aware documentation—including markdown and Mermaid diagrams—for large, multi‑language codebases. It features hierarchical decomposition, recursive multi‑agent processing, configurable token limits, incremental updates, and GitHub‑Pages‑ready output.

LoRA Manager for ComfyUI - A powerful extension for organizing, previewing, and integrating LoRA models with metadata and workflow support.

An enterprise-grade, pluggable agent runtime built on Spring Boot that provides multi-vendor failover, durable long-term task management, and governed digital employees for teams.

A fork of SillyTavern that provides a customizable user interface for LLMs, image generation, and TTS, featuring improved API design and extension capabilities.

Rome is an open‑source “agentic OS” that lets humans and AI agents collaborate in a persistent, self‑hosted environment. It introduces **Rome Apps**—git‑tracked packages containing actions, agents, skills, UI, and data—so capabilities survive beyond a single chat. You can run it locally via Docker/Electron or use the preview Rome Cloud, build apps with the provided SDKs, and share them through an app store. The platform focuses on composable, reusable software rather than just model prompts, positioning itself between persistent‑agent services and personal‑software platforms.

A prompt engineering skill for AI portrait photography that enables consistent style and character recreation across multiple shots by reconstructing professional photographic plans.

An open-source CAN bus toolkit for Flipper Zero and ESP32 that unlocks FSD features, restores banned TLSSC functionality, and provides a steering "nag killer" for Tesla vehicles.

A Python API for interfacing with cars via the CAN bus, enabling the reading of vehicle state and control of steering, gas, and brakes for ADAS development.

A Python-based API and MCP server that allows users to control Xiaomi Mijia smart home devices via code, CLI, or AI agents.

SciPy is an open-source library for mathematics, science, and engineering that provides efficient numerical routines for tasks like optimization and statistics.

SHUO Canvas is a desktop, node‑based canvas that lets creators combine text, images, video and audio with AI generation (via OpenAI‑compatible, RunningHub, ComfyUI, etc.) to build scripts, storyboards, character‑swap videos, 360° panoramas, voice‑overs and 3‑D scenes—all within one visual workflow.