Post 1 in the AI Software House series.

In short: Instead of building a new database or custom framework to run the AI team, I used GitHub itself — issues, branches, pull requests, and Actions — as the backbone. Agent behaviour lives in plain markdown files, not code, so anyone can read and change it. This design means adding a new agent takes minutes, not days.

Every multi-agent AI system needs to answer the same basic questions: where does state live, how do agents communicate, and how does a human stay in control without becoming a bottleneck?

Most implementations reach for a new database, a custom orchestration layer, or a specialised agent framework. I did the opposite. I used GitHub.

Full article in Patreon: https://www.patreon.com/posts/architecture-why-156503054