@lesto/content-search

Client-safe vector search for semantic content discovery.

Installation

npm install @lesto/content-search

Quick Start

createSearch takes the index URL as a string and resolves to a client. The client queries by embedding vector (this package is embedding-free at runtime — generate the query embedding with @lesto/content-embeddings or an embed API):

import { createSearch } from "@lesto/content-search";

const search = await createSearch("/.docks/search-index.json");

// `queryEmbedding` is a number[] (e.g. from POST /api/embed)
const results = search.query(queryEmbedding, {
  limit: 5,
  threshold: 0.5,
});

results.forEach((r) => {
  console.log(r.id, r.score);
});

The client also exposes findSimilar(id, k), getEntries(), and getByCollection(collection).

Features

• Semantic search - Find content by meaning, not just keywords
• Browser-safe - Runs entirely client-side
• Zero dependencies - Lightweight and fast
• Progressive loading - Load index chunks on demand
• Binary quantization - 32x smaller indexes

React Hook

useSearch takes indexPath (not indexUrl) and manages query embedding, debouncing, and keyword/semantic blending internally. It exposes isSearching and isReady (there is no isLoading):

import { useSearch } from "@lesto/content-search/react";

function SearchBox() {
  const { results, search, isSearching, isReady } = useSearch({
    indexPath: "/.docks/search-index.json",
  });

  return (
    <input
      disabled={!isReady}
      onChange={(e) => search(e.target.value)}
    />
  );
}

Documentation

Full documentation at usedocks.dev

License

MIT