67 resources
Ballpoint Breach is a procedural, hand-drawn browser first-person shooter built on Three.js and TypeScript. All assets (arena, weapons, enemies, etc.) are generated through code, no external models. It supports local operation and provides a complete development, testing and construction process, suitable for experiencing unique ink-style FPS gameplay in a web environment.
Infomarchy is a plug-in for Omarchy desktop that converts wallpapers into real-time information panels. It monitors and displays the status, cost, rate limit and machine performance of all running AI agents (such as Claude, Ollama), and supports 7-day heat maps and recent prompt words. The display can be quickly switched through shortcut keys, helping users understand system resources and AI usage at a glance.
Gondolin is a TypeScript-based Linux micro-virtual machine sandbox for safely running AI agent code. It isolates the execution environment through QEMU or krun backend, and controls network and file system access by host policy to prevent data leakage. Supports CLI interaction and programming interfaces, suitable for code execution scenarios that require restricted permissions.
Magnitude is an open source inference server that automatically detects hardware and runs the best local models. It supports offline and private use, and can be seamlessly integrated into mainstream AI agents such as Pi and Cline to achieve localized intelligent development assistance.
InkOS is an AI Agent system for story creation and multi-language translation, supporting the generation of novels, scripts, interactive games and IP content. It unifiedly manages creativity, character setting, memory and review processes through intelligent agents, and provides multiple interactive forms such as Studio, TUI and CLI to help creators efficiently complete the entire process from conception to cross-language delivery.
Atlas is a source code control system designed specifically for coding agents. It supports the use of multiple encoding agents, automatically records checkpoints for each run, associates submissions with corresponding sessions, prompt words, and tool calls, helping users track changes and query agent history in a unified interface.
Provides a set of React components for MDX that supports embedding ASCII-style tables, charts, and flowcharts in Markdown. It needs to be installed based on the shadcn project, and visual data display is achieved through JSX, which is suitable for Web front-end development scenarios.
This is a free AI model availability and latency monitoring radar running on Cloudflare Workers. It regularly evaluates free models on platforms such as Groq and OpenRouter, displays the status and limitations of each provider in real time, and helps users quickly filter and discover the optimal free AI service resources.
Qingyu is an installation-free Markdown and text reading editor specially designed for Windows. It supports offline use, real-time preview, column editing and focus mode. It provides full version and Lite lightweight version. You can enjoy a comfortable local document reading experience without an account.
zframes is an AI-driven personal market terminal framework. Users describe their needs through natural language, and the AI agent automatically generates JSON specifications and renders dashboards containing real-time financial data. It does not require an API key, supports multiple public data sources, and has self-correction capabilities, allowing AI to directly generate front-end interfaces without writing React code.
Open Flow is an open source visual workflow automation platform that supports building and running workflows through CLI and AI Agent. It provides real-time TypeScript type checking, isolated execution environment and self-hosted Control API, allowing users to intuitively edit and monitor processes in Workbench, and is suitable for automation scenarios that require controllability and code transparency.
An open source customer insight workspace that supports search calls, work orders, and Slack conversations. Provides semantic search, company view and MCP endpoint based on Together AI. It has passed Clerk certification and is suitable for building a private intelligence layer.
AppStore Forge is a native Mac application used to convert original application screenshots into store materials that comply with App Store and Google Play specifications. No accounts, uploads or servers required, data remains entirely local. Supports customization of device borders, backgrounds, titles and panorama layouts, and provides WYSIWYG preview and export functions.
Kima Hub is a self-hosted audio streaming platform that brings the Spotify experience to personal music libraries. Supports artist discovery, personalized playlists and podcast subscriptions, and can be integrated with Lidarr and Audiobookshelf. Provides web interface and iOS native applications, and Android is supported through PWA.
MonoCode is a desktop graphical interface tool for coding agents. It integrates multiple AI programming assistants such as Claude Code, Cursor, Grok Build, etc., and provides a unified session management and input interface. The application does not sell tokens and is designed to simplify the experience of using multi-platform coding agents.
Covan is a team-oriented shared AI agent platform that supports self-hosting. Team members can jointly train an AI agent with a unified knowledge base while enjoying private session isolation based on Postgres row-level security. It has knowledge package management, pgvector-based citation answers, automated routine tasks and collaboration functions, aiming to provide intelligent assistants for the entire company, not just engineers.
BeautifulGrid is a high-performance, feature-rich open source React data grid component designed for data-intensive business applications. It is written in TypeScript, supports virtual scrolling, Excel-like cell selection and clipboard operations, provides zero runtime CSS styles and a beautiful interface, and is suitable for web front-end development.
Weight Atlas is an interactive visualization tool for in-depth analysis of LLM model weights. It supports unfolding model tensors into a single canvas, calculating errors, spectral distributions and outliers through actual quantization (INT8/4, FP8), helping developers losslessly compress models and understand their internal structures.