Start a New Project

Build the first scenario as an end-to-end slice. This exposes vocabulary, traceability, and validation issues before the model becomes large.

1. Scaffold

memo init my-device
cd my-device
memo ontology show
memo validate .

The generated project contains:

package.json
memo.package.yaml
memo.lock.yaml
syside.toml
analysis/
  Samples/
    README.md
    01-model-overview.ipynb
    ...
src/
  architecture/
  assurance/
  artifacts/
node_modules/@memoarchitect/ontology/

syside.toml lets SysIDE resolve the project source and the installed ontology as one model.

analysis/Samples contains seven model-independent Jupyter notebooks created by memo init, including charts, an SVG ownership graph, and an HTML/CSV inventory table. Start JupyterLab from the project's analysis directory, then use Analysis → Jupyter Notebooks in Architect to open it.

2. Establish scope

Record:

• intended use and use environment;
• important actors and external systems;
• one operational scenario;
• explicit model boundaries.

3. Add one connected slice

For the scenario, add:

1. a stakeholder need;
2. a measurable system requirement;
3. a logical function;
4. a responsible logical component;
5. a relevant hazard and risk control;
6. a verification case and evidence placeholder.

Use stable identifiers from the beginning:

requirement detectFault : Requirement {
    attribute :>> id = "REQ-001";
    attribute :>> requirementKind = RequirementKind::system;
    attribute :>> name = "DetectDeliveryFault";
    attribute :>> statement =
        "The device shall detect a delivery fault within two seconds.";
}

4. Organize for review

Add architecture definitions under src/architecture, assurance definitions under src/assurance, and model-backed outputs under src/artifacts.

5. Review in Architect

memo validate .
memo-architect dev

Review the slice in context, requirement, allocation, risk, and verification views. Correct the meaning of gaps rather than adding placeholder links.

Then add the next scenario.