//! Parameter types for the `search` verb (ADR-023, ADR-024).
use schemars;
use ;
/// Input for `search` — semantic/hybrid search across entities or notes.
///
/// kind="entity": hybrid search (FTS5 + optional vector) against the entity store.
/// kind="note": hybrid search with salience weighting (ADR-024 pipeline).
///
/// Both pipelines run FTS5 text search + optional vector similarity (if an embedding
/// model is configured), fused via Reciprocal Rank Fusion (k=60).
/// Note search additionally applies salience weighting: score *= (0.5 + 0.5 * salience).
///
/// Soft-deleted and superseded notes are excluded from results.
///
/// Examples:
/// Find entities: {"kind":"entity","query":"FlashAttention memory efficient attention"}
/// Find notes: {"kind":"note","query":"LoRA fine-tuning parameter efficiency","limit":5}