Hey HN,<p>Michael and Scott here. We’re open-sourcing an interactive murder mystery featuring LLM-driven character agents. Solve the mystery by finding clues, taking notes, and interrogating agents. They all have distinct motives, personality, and can impact the game in different ways (attacking you, running away, etc). Try it out, it’s pretty fun!<p>We’re also open-sourcing the framework that we used to make and refine the agents. The goal is to create an intuitive interface for storytellers to create, debug, and test game agents. We then take those game agents and expose an API beyond just chat - such as actions, player guardrails, emotional queries, etc.<p>We’re not done yet - there are a lot more features coming on the way: scenario-based agent evals, agent-storyline consistency management, automatic agent generation, etc.<p>We would love to hear your feedback.<p>Thanks!<p>[0] <a href="https://github.com/mluogh/grontown">https://github.com/mluogh/grontown</a><p>[1] <a href="https://github.com/mluogh/eastworld">https://github.com/mluogh/eastworld</a><p>[2] <a href="https://discord.gg/qRaWE2jp" rel="nofollow noreferrer">https://discord.gg/qRaWE2jp</a>
This is a cool demo! It does a good job hinting at how immersive the experience can be.<p>I'm excited about large language models for gaming. The first thing I tried with ChatGPT was to run a text-based RPG game, and it was fun the very first attempt. I imagine we'll have (a text-based) holodeck/Westworld soon enough.<p>In the demo I ran out of tokens pretty quickly, after talking to 3 of the characters. I wonder how the economics of this works - I'm guessing token markup?<p>I wrote a library for structured inference in-browser[1] and recently started building a murder-mystery demo as well[2] (very rough WIP, not yet playable).<p>The library might make a good backend for your project. Local llm in-browser means it's way easier to distribute since there's no need to install anything. Structured inference lets you get consistent responses out of even smaller models so you can always get a conversation, or an action json. And of course, no token limits since it runs client-side.<p>[1] <a href="https://github.com/gsuuon/ad-llama">https://github.com/gsuuon/ad-llama</a>
[2] <a href="https://github.com/gsuuon/ad-llama/tree/murder-mystery-solidjs/example/vite-demo/murder">https://github.com/gsuuon/ad-llama/tree/murder-mystery-solid...</a>
I spent a few minutes chatting around - this is fun. I tried to get Amelia to go fetch me a shovel to unbury the body - but she did not collaborate much 10/10.
Imagine something like Baldur's Gate 3 with AI generated, infinite stories and characters in those stories. Including voice acting.<p>I think you're having the right idea.
I have to laugh.<p>First, the j_l button on my _eyboard is not functional so I have to sort out how to words without it.<p>Second, the characters, when faced with vague questions, will give away pieces of the plot before I discovered them. For example, a doctor let slip the manner of death before I ever saw the body of the deceased. Someone, when questioned vaguely about the "night of the party" first said the deceased <i>died</i> and then when questioned more closely said <i>collapsed</i>.<p>There's a component of information discovery here that should be rolled out to each character. The doctor should not have information before the discovery of the information.<p>I'm going to laugh if this is a "Murder on the Orient Express" ending.