This project provides a ChatGPT plugin for the YunZai-Bot, enabling AI-powered chat functionalities within a QQ bot environment. It targets users of the YunZai-Bot who want to integrate advanced large language models like GPT-3.5, GPT-4, Claude, and Gemini, offering features such as conversation memory, AI personality tuning, and direct voice synthesis.

How It Works

The plugin supports multiple modes of interaction with large language models, including direct API calls (OpenAI, Azure, etc.) and a browser-based fallback. It leverages the Conversation API for stateful interactions and offers a "smart mode" that allows the bot to perform actions like muting users, changing group names, and performing web searches, requiring additional API integrations for these capabilities.

Quick Start & Requirements

• Install via git clone into the YunZai-Bot plugins directory, followed by pnpm i.
• Requires Node.js >= 18.
• Configuration is managed via plugins/chatgpt-plugin/config/config.json or a backend panel.
• Official documentation and setup guides are available.

Highlighted Details

• Supports a wide array of LLMs including GPT-3.5/4, Claude, ChatGLM, Tongyi Qianwen, and Gemini.
• Offers features like AI personality tuning, role-playing, and direct voice synthesis (VITS, Azure).
• Includes "smart mode" for bot actions and "pseudo-person mode" for controlled random chat triggers.
• Supports file reading for specific models and modes.

Maintenance & Community

The project is actively under refactoring with a v3 branch being the primary focus. Community support is available via QQ groups.

Licensing & Compatibility

The project's licensing is not explicitly stated in the README, but it references and uses other open-source projects. Compatibility for commercial use or closed-source linking is not detailed.

Limitations & Caveats

The README notes that documentation may be partially outdated due to rapid development. Some features, like GPT-4 via API3, were reported as potentially blocked by OpenAI. The project is undergoing significant refactoring, with the v2 branch receiving minimal maintenance.