A development framework that connects large models to databases, tools, and external systems, allowing you to quickly build AI applications without reinventing the wheel.

An open-source AI knowledge base engine that lets you upload documents and chat with them, getting accurate answers without hallucinations.

A lightweight command-line coding agent by OpenAI that helps you write code, fix bugs, and manage files directly in your terminal without needing a browser.

An open-source project simulating an AI hedge fund, featuring AI avatars of 19 investment masters like Buffett to analyze stocks and make decisions, for learning and entertainment only.

An open-source AI assistant that thinks proactively, operates computers, and connects to WeChat, helping you handle complex tasks with long-term memory.

An open-source AI app monitoring and debugging tool that helps you trace LLM calls, manage prompts, and evaluate performance for transparent and controllable development.

An enhancement plugin for OpenAI Codex CLI that adds structured workflows and agent commands to make AI coding more organized and efficient.

A Python framework that makes it easy for developers to build multi-agent collaboration systems, managing task allocation, tool usage, and safety checks across major large models.

This tool enables AI coding assistants (like Claude Code) to understand your entire codebase via semantic search, pinpointing relevant code without feeding the whole project to AI, saving money and boosting efficiency.