Update documentation

[kostapetan]

Documentation is grossly outdated and we don't even have proper demo/explanation of the github flow, nor the abstractions that surfaced while building this.

I suggest we work on documenting:

• [x] How to run this locally
• [x] Reasoning and rationally behind the project and the AiAgents
• [x] Event flow for the GH dev agents

[katarinasvedman]

I miss a picture of the flow in the demo. And some clarification that the implementation is ONE example but the aim is to find patterns to reuse for other use-cases so this eventually becomes a "framework" to accelerate building agents. What building-blocks does this "framework" provide?