A reusable AI agent layer that defines roles, rules, concepts, processes, and conventions for development according to the Ergonomic Approach (EA).
Recommended integration structure:
<project-root>/
├── .ai/
│ ├── ergo/ # git submodule → this repository pinned to a commit
│ └── project-local/ # project-specific extensions and overrides
└── ...
- INDEX.md — start here (entry point and artifact index).
- agents/ — agent roles and collaboration model.
- skills/ — executable skills and procedures.
- concepts/ — formalized concepts (operational definitions and specifications).
- conventions/ — shared conventions and formatting rules.
- conventions/ea-principles.md — EA principles as top-level review criteria.
- checklists/ — review checklists aligned with EA principles.
- bootstrap/ — initial project bootstrap templates to activate the framework.