Autonomous, sentient-like AI bots powered by local LLM inference
This is the GitHub Pages website for the Protocol Luna organization.
An open-source multi-platform AI assistant ecosystem. The brain service (Emerald) connects to platform adapters (Discord via Jade, Matrix via Pixieglow) and an LLM gateway (Sapphire) backed by llama.cpp (Krystal). A Markov chain service (Ruby) adds context-free spontaneity.
| Service | Role |
|---|---|
| Krystal | LLM inference via llama.cpp |
| Sapphire | LLM gateway with emotion classification |
| Emerald | Brain & behavior engine |
| Jade | Discord adapter |
| Pixieglow | Matrix adapter |
| Ruby | Markov chain service |
# Serve locally
python3 -m http.server 8000The site is deployed via GitHub Actions to GitHub Pages.