🐧 Proposal for Linux contributions

[arslan-charyyev]

Good day, @mrexodia !

First of all, allow me to express my sincere gratitude for the amazing work with x64dbg over the years. Second, please forgive me for the wall of text below, but I thought it's important to give the full context before I submit my proposal.

I've enjoyed using x64dbg for debugging and analysis while Windows was my primary daily driver. But since switching to Linux I could not find an adequate alternative. What I ended up doing is developing a personal plugin for EDB debugger since it was closest in spirit to x64dbg (quite snappy, doesn't freeze the system when opening large binaries, extensible with plugins). Now, the development was entirely driven by AI agent (Claude Opus 4.8 specifically) and it was able to implement all of the features that I needed: string search, cross-references, caching, and even MCP support so that features are accessible not only to humans via Qt frontend, but also to agents via embedded HTTP server). I only remained in the loop to define tasks, roadmap, smoke test, and to keep the plugin's architecture consistent with best practices to avoid generating sloppy code.

Now, while I can keep going with this setup, I though that perhaps this is the wrong path long-term. It would be far preferable to use x64dbg as the platform instead of re-inventing the wheel. Luckily I saw that there was some groundwork was done by @3rdit . I kept monitoring the situation but I haven't seen any progress being made recently. And this brings me to the question/proposal I wanted to submit to you.

1. Would you be willing to accept contributions authored by AI agents to port x64dbg to Linux with full feature parity? I am starting my proposal with this question because many in the open-source space do not welcome such contributions (Zig programming language is a notable example). My intention is to use only frontier models for such contributions - Opus 4.8 @ Extra thinking effort or Fable 5 once it's available.

2. If you are open to such contributions, how would you like the development to proceed? I would start by creating a new project board in GitHub to track the backlog. GitHub issues can be used as a platform for technical discussions on non-obvious design choices.

[3rdit]

Hi.

You are welcome to make AI generated contributions for the Linux port provided they fit the quality standard of the project - the code should be tested and correct. mrexodia doesn't really have an involvement in the development of the Linux port other than ensuring it is of standard and works, therefore, I would likely be the one to do the initial code reviews of any PRs you make.

I have had a busy month, hence why the development has slowed down. Additionally, this project as a whole doesn't really use the project board for backlog tracking, so I don't think it's something we want to do. The logical next steps for the Linux port is to create more of the cross components and set up the interface to match the Windows further.

If you're serious about the help of the project you can collaborate with us via the Discord server, my tag is azulx.

[arslan-charyyev]

Thank you very much for the reply, @3rdit ! I'm happy to hear that AI contributions are welcome. The requirement of high quality and test coverage is perfectly aligned with my intentions. I will reach out to you via Discord soon.

[3rdit]

Join the official server here: https://discord.x64dbg.com/

[arslan-charyyev]

Thank you! I submitted my application to join the server.