Generated from
smart-base/skills/content/authoring-who-smart-guidelines/recommendation-scenarios.md— do not edit here.
recommendation-scenarios
Step 2 of recommendation-extraction. That
skill’s Part 2 holds the judgement (the shape of a scenario and the
reuse-first precedence); this one is the mechanical step that applies it.
Input and output
- Input:
input.schema.json - Output:
output.schema.json
In: the L1 graph from recommendation-candidates
(or an authored L1 YAML), the target guide’s vocabulary, optionally the generic
personas, and the agent’s mapping.
Out: scenario-candidates.l2.json, one candidate per recommendation, with
warnings for what could not be determined. Every candidate stays candidate
until a person reviews it.
Tool
l1-recommendation-scenarios (smart-base/tools/l1-recommendation-scenarios/l1-recommendation-scenarios.ts):
bun run smart-base/scripts/l1-recommendation-scenarios.ts --l1 <l1.json|yaml> --vocabulary <guide.json> [--generic <generic.json>] --mapping <mapping.yaml> --out <dir> [--check]