Goal-Oriented-Agents is a project that aims to develop multi-agents systems that can learn to achieve goals in a complex environment. And the core part is borrowing from Generative Agents created by @joonspk-research.
- clone the repository to your local machine
- make sure you have installed Node.js
- run
npm installin the root directory of this project - create a .env file in the root directory according to the key-value pairs in the .env-template (provide your AK and SK)
- run
npm run testto start simulation, you can adjust the LOG_LEVEL in the .env file to see more detailed or more concise log (error, warn, info, http, verbose, debug) - you can check the simulation results in the
agentsdirectory (next_action.jsonandglobalConversation.json) - any error message will be documented in the combined.log file, you can check it to see what went wrong.
- add task assignment part
- connect to the frontend and unity