This announcement formally introduces DexSQL, accessible via the following link: https://github.com/LSPosed/dexsql-release. The project facilitates the analysis of APK dex files using SQL, specifically for deobfuscation and the identification of hook points. Compared to tools such as JADX and similar counterparts like ASC, DexSQL offers significantly superior performance and memory consumption, as illustrated in the attached figures. Furthermore, it supports MCP servers, enabling AI agents to perform analysis and develop LSPosed modules efficiently while minimizing token usage. DexSQL will be integrated into LSPosed in the future, allowing users to utilize the libxposed API to locate hook points and support cross-version deobfuscation.

该公告正式介绍了 DexSQL，可通过以下链接访问：https://github.com/LSPosed/dexsql-release。该项目利用 SQL 分析 APK 的 dex 文件，专门用于去混淆和识别挂钩点。与 JADX 及 ASC 等类似工具相比，DexSQL 的性能内存占用显著优越，如附图。此外，它支持 MCP 服务器，使 AI 智能体能够高效地执行分析并开发 LSPosed 模块，同时最大限度地减少词元使用量。未来，DexSQL 将被集成到 LSPosed 中，允许用户使用 libxposed API 来定位挂钩点并支持跨版本去混淆。