This was made collaboratively by me directing coding agents at the binary, using Ghidra MCP extensively, diassembly and also dynamic analysis with an emulator. I don't have a writeup of the process but it was definitely not fully automatable (I wish though). I might prepare a blog post with transcripts and session history and things I learned along the way.
Broad takeaways:
- Ghidra MCP is not a silver bullet. Lots of opportunities for mis-decoding especially on older instruction sets (e.g. conflating code + data), which requires user input to flag data layout/structs.
- Agents still need a lot of user direction otherwise the RE production is just kind of a random walk. With Z80 it's decent at reading code but I expect that it has much worse performance than reading x86 or ARM for instance. The TI-84+ has a bunch of hardware quirks as well.
- GPT 5.5 is better than Opus 4.8 at RE. Opus 4.8 loves plausible-sounding RE'd logic without any checking. The gold standard is actually dynamically executing the binary and comparing the logic against the prose.
- Maintaining consistency in style and prose is a PITA across the wiki. Hard to reconcile prose <-> code. Can be somewhat mitigated by agent loops.
Was also in discussions with people in the TI calculator programming space who helped provide guidance as well. We previously did not have a catalogue of every subsystem in TI-OS yet alone most subroutines in the OS.
It's highly likely that the original implementation language was assembly. The code is very idiomatic.
Regarding source build, I think reverse engineering it to the point where you can reconstruct the source is possibly legally problematic, so I don't plan to do this, but maybe for certain subsystems like MathPrint (equation display) which was especially fun to RE. I have a PR up for it and it will be live at
The plans are heavily subsidized by the AI companies so I didn't end up needing to do API usage or buy another subscription. I have ChatGPT Pro and Claude Code Max.
spends $100/mo (ChatGPT Pro) and $100/mo or $200/mo ("5X" and "20X", respectively) for AI tools.
So between $200-300/mo or $2,400-3,600/yr.
is that what you mean by "heavily subsidized by the AI companies"? What a joke.
No, your silly research is subsidized by middle America through higher costs of utility bills, heavy extraction and depletion of water tables, use of stolen or misappropriated land for AI specific datacenters, and the continued pollution of the world.
Broad takeaways:
- Ghidra MCP is not a silver bullet. Lots of opportunities for mis-decoding especially on older instruction sets (e.g. conflating code + data), which requires user input to flag data layout/structs.
- Agents still need a lot of user direction otherwise the RE production is just kind of a random walk. With Z80 it's decent at reading code but I expect that it has much worse performance than reading x86 or ARM for instance. The TI-84+ has a bunch of hardware quirks as well.
- GPT 5.5 is better than Opus 4.8 at RE. Opus 4.8 loves plausible-sounding RE'd logic without any checking. The gold standard is actually dynamically executing the binary and comparing the logic against the prose.
- Maintaining consistency in style and prose is a PITA across the wiki. Hard to reconcile prose <-> code. Can be somewhat mitigated by agent loops.
Was also in discussions with people in the TI calculator programming space who helped provide guidance as well. We previously did not have a catalogue of every subsystem in TI-OS yet alone most subroutines in the OS.
Regarding source build, I think reverse engineering it to the point where you can reconstruct the source is possibly legally problematic, so I don't plan to do this, but maybe for certain subsystems like MathPrint (equation display) which was especially fun to RE. I have a PR up for it and it will be live at
https://siraben.github.io/ti84p-re/mathprint
So between $200-300/mo or $2,400-3,600/yr.
is that what you mean by "heavily subsidized by the AI companies"? What a joke.
No, your silly research is subsidized by middle America through higher costs of utility bills, heavy extraction and depletion of water tables, use of stolen or misappropriated land for AI specific datacenters, and the continued pollution of the world.
> The big picture
> The structural reverse-engineering is comprehensive (every subsystem mapped, both cross-page mechanisms resolved ...
> Confidence summary / open items
Probably an LLM wrote the docs.
> (the GhidraMCP plugin reconnects for interactive work)
Probably LLM+Ghidra for the actual RevEng. Ultimately does it matter if the end product is works though