A murder mystery game built on an open-source gen-AI agent framework
Hey HN,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!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.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.We would love to hear your feedback.Thanks![0] https://github.com/mluogh/grontown[1] https://github.com/mluogh/eastworld[2] https://discord.gg/qRaWE2jp
Users praised the immersive experience and the use of LLM, particularly noting the AI-generated stories in Baldur's Gate 3 and the realistic character memory. However, there were concerns about token limits, a keyboard issue, plot spoilers, and the need for personal details. The demo works well on laptop GPUs, but there were issues with character behavior and information discovery. One user had a negative experience with a character, while another mentioned a problem with a chat feature. A library suggestion was made, and one comment was non-contributory.
Users reported that the product depletes tokens too fast and has difficulty solving logic puzzles. There's dissatisfaction with a character named Amelia for not collaborating, and concerns about characters disclosing plot details too soon. Additionally, users are confused about the necessity of providing personal details for the demo.
