Reverse-engineer a 1997 PC game and make it run on modern hardware
Claude Code got an old DirectX game running on modern hardware, decompiled much of it, built a headless parallel runner, and now writes full level guides with screenshots.
A user had ChatGPT with Computer Use control a Pokémon Silver emulator, picking a starter and starting a playthrough, though usage limits were consumed by setup and repetitive dialogue.
A user had ChatGPT with Computer Use control a Pokémon Silver emulator, picking a starter and starting a playthrough, though usage limits were consumed by setup and repetitive dialogue.
The user ran Pokémon Silver in an emulator and let ChatGPT, which they call Lyric, control it through Computer Use. They played alongside it instead of having it only give advice. The agent chose Totodile as its starter and named it Riptide. They then began setting up a playthrough the user could stream through OBS.
It worked well enough to start the game and pick a starter. Setup, navigation, repetitive dialogue and simple movement used up a lot of the Computer Use allowance before real play began.
Plus-plan Computer Use limits drain quickly on repetitive game inputs, and the post says it doesn't show how much allowance is left or when it resets.
Open the emulator on my desktop with [game name and ROM already loaded], pick [starter or character choice], and play for about [number] minutes, advancing dialogue and exploring. Keep a short log of progress and stop at [milestone] or when you reach the time limit. Don't download anything or change system settings without asking me first.
Claude Code got an old DirectX game running on modern hardware, decompiled much of it, built a headless parallel runner, and now writes full level guides with screenshots.
ChatGPT Agent Mode opens a real browser and books appointments by clicking through the booking site; the post gives a setup guide.
A developer used ChatGPT Agent mode to extract revenue growth, free cash flow and net cash from 10-Q filings with source URLs, feeding a multi-agent valuation and sentiment workflow.