I've been working on a remote GUI debugger as part of my personal security research. It’s currently a work in progress, so only a subset of features is implemented at this stage.
For the sake of my own learning, I’ve decided to implement a custom analysis engine from scratch rather than relying on existing frameworks like Frida, LLDB, or GDB.
If you’re interested, I’d love for you to give it a try and let me know what you think!
For example, hardware watchpoints on iOS are significantly faster than those from debugserver. This is just one example, but I believe existing frameworks are not necessarily optimized for reverse engineering.