Contributing

Thank you for your interest in contributing! This project was open-sourced to build a community of people interested in building our own personalized Agents.

Contribution Wishlist

The primary categories of ways to contribute are:

1. [Data Extractors] Implementation of data extractors to pull in new personalization data. You can find a specific wishlist at data/extract/README.md
2. [Data Transformation]
3. [Data Storage]
4. [Agent Architectures] Implementation of new MirrorAgent architectures. We have provided one sample implementation of a langchain conversational agent using tools connected to a ChromaDB, but there is a big opportunity for experimentation on what works best here. See the existing agent definitions in mirror_agent/agents
5. [UX Experiences] Implementation of fun, interesting, and/or useful experiences that use these personalized agents. The project was initially kicked off with the ability to interact with a MirrorAgent through either text->text chat or text->voice chat, but we are interested in building lots of additional UIs utilizing the core MirrorAgent. See voice/README.md
6. [Configuration] As these agents become increasingly complex, managing their configuration will become harder. We are at the design phase for how to approach this and welcome ideas.

Guidelines

TODO