score_cognitive_local

Compute and persist cognitive-difficulty scores from a VALIDATED, deterministic READABILITY metric - the local-first, 0-credit, no-AI path (SC 3.1.5). English uses Flesch-Kincaid; every other language uses LIX, both normalised to difficulty 0-100 (higher = harder) with a per-language threshold. Writes cognitive_score for the scoped keys; a key enters the reading_level_high queue once its score >= the project threshold. Prefer this over analyze_cognitive (billed AI) for difficulty scoring. Scope with key_uuids and/or namespace_uuid (omit both = whole project). Returns the scoring summary.

Paramètres

Nom	Type		Description
project_uuid	string	optionnel	Project UUID. Required when multiple projects are configured (SONENTA_PROJECTS); defaults to the single configured project otherwise.
key_uuids	array	optionnel	Restrict scoring to these keys. Omit to scope the whole project (or a namespace).
namespace_uuid	string	optionnel	Restrict scoring to one namespace by its UUID.
overwrite	boolean	optionnel	Re-score keys that already have a cognitive_score (default false = only score unscored keys).

Exemple

{
  "tool": "score_cognitive_local",
  "arguments": {}
}