Kuna: Decompiler Development in the Age of Coding Agents

#1

Post by matt_d » Thu, Jul 30, 2026, 2:41 AM UTC

Kuna: Decompiler Development in the Age of Coding Agents
noelo.org

Re: Kuna: Decompiler Development in the Age of Coding Agents

#2

Post by fishfasell » Thu, Jul 30, 2026, 4:02 AM UTC

It would be cool to see agentic interpretation of function and variable names. It can see and track the flow of data a lot faster than a human can so if given some context, maybe it could synthesize names for them.

Re: Kuna: Decompiler Development in the Age of Coding Agents

#3

Post by DobarDabar » Thu, Jul 30, 2026, 4:17 AM UTC

[dead]

Re: Kuna: Decompiler Development in the Age of Coding Agents

#4

Post by snoopen » Thu, Jul 30, 2026, 5:06 AM UTC

fishfasell wrote: post #2

It would be cool to see agentic interpretation of function and variable names. It can see and track the flow of data a lot faster than a human can so if given some context, maybe it could synthesize names for them.

I've used Claude with Ghidra MCP and had good success. It correctly identified and named MD5 functions, file IO functions, and the whole chain from a UI dialogue function. And when prompted it named all the interesting variables too.

Re: Kuna: Decompiler Development in the Age of Coding Agents

#5

Post by hackcasual » Thu, Jul 30, 2026, 5:30 AM UTC

snoopen wrote: post #4

fishfasell wrote: post #2

It would be cool to see agentic interpretation of function and variable names. It can see and track the flow of data a lot faster than a human can so if given some context, maybe it could synthesize names for them.

I've used Claude with Ghidra MCP and had good success. It correctly identified and named MD5 functions, file IO functions, and the whole chain from a UI dialogue function. And when prompted it named all the interesting variables too.

I was working on adding better WASM support to ghidra, but now I can just point Claude to binaryen

Re: Kuna: Decompiler Development in the Age of Coding Agents

#6

Post by bpavuk » Thu, Jul 30, 2026, 6:04 AM UTC

hackcasual wrote: post #5

snoopen wrote: post #4

Earlier quoted context omitted.

I've used Claude with Ghidra MCP and had good success. It correctly identified and named MD5 functions, file IO functions, and the whole chain from a UI dialogue function. And when prompted it named all the interesting variables too.

I was working on adding better WASM support to ghidra, but now I can just point Claude to binaryen

I wonder, how many such interesting and universally beneficial developments did emergence of tools like Claude Code kill?

Re: Kuna: Decompiler Development in the Age of Coding Agents

#7

Post by saidnooneever » Thu, Jul 30, 2026, 6:33 AM UTC

this is really amazing work thank you and in my opinion (as its explained) a very good example of how to use AI powered development and research to advance the tools we have. decompilation is really hard, and better algorithms for it are very valuable contributions for many areas of tech.

Re: Kuna: Decompiler Development in the Age of Coding Agents

#8

Post by nneonneo » Thu, Jul 30, 2026, 6:39 AM UTC

hackcasual wrote: post #5

snoopen wrote: post #4

Earlier quoted context omitted.

I've used Claude with Ghidra MCP and had good success. It correctly identified and named MD5 functions, file IO functions, and the whole chain from a UI dialogue function. And when prompted it named all the interesting variables too.

I was working on adding better WASM support to ghidra, but now I can just point Claude to binaryen

Well, that's too bad; as the main developer of https://github.com/nneonneo/ghidra-wasm-plugin it would have been really nice to get better support for all the new features that have been added into the spec over the last few years.

I also think that, for larger projects, it's still useful to have real decompilation support; for example, Il2CppDumper works in tandem with ghidra-wasm-plugin to enable decompilation of Unity Il2Cpp projects, which are often enormous (100MB wasm files are not uncommon) and for which the symbols + types are invaluable.

Re: Kuna: Decompiler Development in the Age of Coding Agents

#9

Post by ChickeNES » Thu, Jul 30, 2026, 7:14 AM UTC

snoopen wrote: post #4

fishfasell wrote: post #2

It would be cool to see agentic interpretation of function and variable names. It can see and track the flow of data a lot faster than a human can so if given some context, maybe it could synthesize names for them.

I've used Claude with Ghidra MCP and had good success. It correctly identified and named MD5 functions, file IO functions, and the whole chain from a UI dialogue function. And when prompted it named all the interesting variables too.

I'm not sure you even need the MCP tbh, I just have Claude use pyghidra scripting.

Re: Kuna: Decompiler Development in the Age of Coding Agents

#10

Post by ur-whale » Thu, Jul 30, 2026, 9:21 AM UTC

Not sure why someone would want to use an LLM to build a new decompiler instead of training an LLM to BE a decompiler.

Especially given the fact that you have an infinite training set to train that LLM from (compilers can generate as much training data as you could possibly want).