central-casting
Overview and contents.
A public demo of a method for turning one open conversation thread about a project into a structured, multi-session system: work lanes, system surfaces, task home folders and schemas. The operating layer is Cursor.
What is here
| Page | What it is |
|---|---|
| Demo | The method as eight cards, served through GitHub Pages. |
| Walkthrough deck | The method as a short slide deck, the primary walkthrough. |
| Walkthrough, reading version | The same method written as eight steps in prose, with the example pages linked at each step. |
| Reference | The problem, the glossary, a worked example, the architecture, concrete moves and objections. |
| In practice | The real catalogue, the tooling layers and how the aimez.ai research program runs on this method. |
The sanitized example set
The example/ folder carries a sanitized schema set the walkthrough points at. It holds no private project content.
| File | Role |
|---|---|
| 00, the orchestrator | The zero element that routes work and reconciles state. |
example/actor_catalogue.yaml | The example work lanes. |
| system surfaces | The file surface classes. |
example/work_home_schema.yaml | The task home folder. |
example/checkpoint_policy.yaml | Checkpoints as the unit of memory. |
| a step log | A real checkpoint history. |
How to use it
Open the walkthrough deck for the quick version, or read the reading version and open the matching example page at each step. The demo page presents the same eight steps in order.
Honest scope
The method is recent and still maturing. It reduces context loss across long, multi-session work, and it depends on care. This repository is a teaching demo, not a framework to install.