peytonshields

This is something we've been working on and are planning to release a "decision" update to the game which should allow for multi-step, configurable options to choose if the LLM actually gets to contribute to the current world / chat. There's a lot of trial and error involved and we're all ears, if you have ideas we'd love to hear them! We actively monitor our discord https://discord.com/invite/theinterface

We've only just begun! Max and I started building this about 1.5 months ago and are planning to ship a torrent of updates for the foreseeable future! Eventually it will be much more open/explorable world

Thanks for the comment! We're working towards using the game's own simulation data (from Unity) to feed back into your game's agents. We hope this will prove less noisy than speech / real-world instrument data, allowing the AI to learn more effectively with new data every time you play

Fixed! Thanks for flagging!

Coming soon! If you join our discord happy to debug live, we have the build for it but figuring out some libgtk dep issues with Tauri

This is the goal! We're working hard to give the AI more spatial "world" awareness with bespoke decision loops

Absolutely! Max and I were huge Civ fans and always tried to make the game AI deviate from its programmed strategy. We also believe you can get some really interesting story arcs by adjusting parameters like temperature and how context is presented. Some of the things you'll notice in the game is we have a no-holds barred approach -- you can fully modify system prompts and adjust how the LLM interprets the state of the world.

The loop is all about adapting, experimenting, and seeing which combinations work and which don't. Right now it's designed around mini-games which can have a different goal per game -- as a quick example I'm currently building agent tic-tac-toe but hidden trapdoors and power-ups