GalQQ enhances the QQ mobile application by integrating a Galgame-style three-option quick reply system directly into the chat interface. It targets QQ users seeking a more engaging or efficient way to respond, offering intelligent, context-aware replies powered by AI or pre-defined local dictionaries. The primary benefit is a more dynamic and convenient chat experience, allowing users to select replies with a long press on message bubbles.

How It Works

This project is an Android module developed using the LSPosed framework, designed to hook into and modify the QQ application's behavior. It intercepts message display and input actions to inject a custom UI element presenting three distinct reply choices. These choices are dynamically generated either by querying an AI service (like OpenAI) based on the conversation context or by retrieving suggestions from an integrated local dictionary, providing both online intelligence and offline utility.

Quick Start & Requirements

• Installation: Download the APK from the Releases page, install it via the LSPosed Manager, enable the module, and restart the QQ application.
• Prerequisites: Android 10.0+ (API 24+), LSPosed framework (recommended) or EdXposed, QQ version 9.0+.
• Links: Official GitHub repository: https://github.com/yiyihuohuo/GalQQ.

Highlighted Details

• Core Feature: Three-option quick reply system for QQ messages.
• AI Integration: Utilizes OpenAI API for contextually relevant, AI-generated reply suggestions.
• Local Dictionary: Supports an internal dictionary for offline reply generation.
• Customization: Extensive configuration options including system prompts, API settings, temperature, and message filtering.
• User Interface: Includes a dedicated settings screen, AI log viewer, and AI monitoring panel.
• Technical: Compatible with QQNT architecture, built with modularity in mind.

Maintenance & Community

The project is hosted on GitHub and encourages community contributions via standard pull requests. A QQ group (859142525) is available for support and discussion. Commit messages follow conventional types like feat, fix, and docs.

Licensing & Compatibility

The project is licensed under the Apache License 2.0. However, it explicitly states it is for "learning and research purposes only" and not for commercial use. Users are warned that using this module may violate the QQ user agreement, and they assume all risks.

Limitations & Caveats

Potential compatibility issues may arise with certain QQ versions. AI-generated replies can occasionally lack contextual accuracy. There is a reported risk of memory leaks after extended periods of use.