Getting Started

Start from a local copy of the full Arketype repository. Python 3.10 or later is recommended; basic browsing and the included HTML catalogue do not require Python, internet access, or a database.

1. Browse the visual catalogue

Open catalogue/index.html from the extracted repository. Search by keyword, filter by family or structural pattern, and inspect the labeled diagrams. Each archetype card links to a descriptive JSON file; use-case cards link to their proposed composition recipes. The catalogue is an offline generated projection.

A printable edition is included as catalogue/Arketype_Visual_Catalogue_v1.4.0.pdf. It is optional: HTML is the primary interface.

2. Inspect the library

From the repository root:

python arketype.py list
python arketype.py search "protocol"
python arketype.py show procedure-protocole
python arketype.py profiles
python arketype.py profile suivi-employe
python arketype.py recipes
python arketype.py recipe troubleshooting-systeme

The CLI uses stable slugs (for example etat-systemes and dossier-suivi). Reader-facing text is in English. Existing French-derived slugs and machine property names are retained as stable identifiers to preserve compatibility.

3. Preview an instantiation

python arketype.py create   --type diagnostic-causal   --slug lab-diagnostics   --title "Lab Diagnostics"   --parent .   --dry-run

The dry run reports the intended destination, archetype and design plan. It writes nothing.

4. Create a new, blank Kristal

Run the same command without --dry-run. Arketype asks the bundled Authoring Kit to initialize an independent workspace; it copies optional structure proposals into design/, verifies the local workspace, and refuses to overwrite an existing destination.

The generated template is empty, not a populated diagnostic knowledge base. Before adding real data, define domain identity, provenance, privacy constraints, semantic ownership, logical contracts and validation criteria.

5. Preview a composition proposal

python arketype.py compose   --recipe troubleshooting-systeme   --slug diagnostics-plan   --title "Diagnostics Plan"   --parent .   --dry-run

This produces a preview of components and unresolved responsibilities. It never fuses independently owned canonical states.

6. Validate

python -m unittest discover -s tests -v
python arketype.py validate --check-kit

To regenerate the HTML after updating source JSON:

python arketype.py build-html

Desktop launcher (Windows)

Double-click Lancer_Arketype.pyw with Python 3.10+ and Tkinter. It opens one application with Home, Catalogue, Create Kristal, and Developer Ops. Search the catalogue, choose an archetype or use-case profile, and send it directly to the creation form. Preview a workspace before creating it.

Developer Ops offers local list, search, show, profile, recipes, validate, validate --check-kit, test, inventory, and build-html operations. Enter help for the exact allowlist. There is no arbitrary shell, remote publication, or activation. Long-running operations run in the background so the window stays responsive. The launcher uses the supplied logo and brand color #1E6864; no logo is embedded in the README.

Native Windows testing remains a separate validation step; headless GUI tests have been performed on Linux.

Next: Working with Templates · CLI Reference · Security and Boundaries