[TERMINAL · SKILLS]
> mounting /skills...
> indexing skill manifests...
> linking agents: claude · codex · gemini · cursor
> ready.
[░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 0%
Use Cases/Build E-Commerce Product Search

Build E-Commerce Product Search

Build a product search engine with full-text search, faceted filtering, typo tolerance, synonym matching, search analytics, and personalized ranking for e-commerce storefronts.

Development#typescript#tsconfig#type-checking#type-safety#javascript-migration
Works with:claude-codeopenai-codexgemini-clicursor

Skills stack · 5 skills

Avg quality 96/100·All SAFE
>

typescript

v1.0.0

Sets up and repairs TypeScript projects: picks tsconfig.json settings for a Node.js service, an npm library or a bundled web app, runs type-checking in CI, and replaces `any` and unsafe casts with types the compiler can verify. Use when someone asks to "set up TypeScript", "fix this tsconfig", "fix a TypeScript error", "upgrade to TypeScript 6 or 7", "migrate JavaScript to TypeScript", "publish types for a package", "run .ts files in Node" or "make this type-safe". Covers the changed defaults in TypeScript 6.0 and 7.0, module and moduleResolution, strict mode, project references, declaration output, narrowing, satisfies and common compiler errors.

93/100 quality
7.30× impact
SAFE· stale score
View skill
>

redis

v1.1.0

Redis is an in-memory data store used as a cache, session store, message broker and counter service. Use when adding a cache layer, storing sessions, rate limiting an API, running pub/sub or Streams with consumer groups, taking a distributed lock, building a leaderboard, or configuring Redis for production (persistence, maxmemory, eviction).

93/100 quality
1.79× impact
SAFE· stale score
View skill
>

postgresql

v1.1.0

Assists with designing schemas, writing performant queries, managing indexes, and operating PostgreSQL databases. Use when working with JSONB, full-text search, window functions, CTEs, row-level security, replication, or performance tuning. Trigger words: postgresql, postgres, sql, database, jsonb, rls, window functions, cte.

100/100 quality
5.15× impact
SAFE· stale score
View skill
>

hono

v1.1.0

Hono is a small web framework built on Web Standards (Request/Response) that runs unchanged on Cloudflare Workers, Deno, Bun, Node.js, AWS Lambda and Vercel. Use it when the user asks to build a Hono API, add Zod validation, JWT auth, CORS or middleware, use the type-safe RPC client (hc), render JSX, stream responses, or deploy one codebase to several runtimes.

100/100 quality
6.20× impact
SAFE· stale score
View skill
>

zod

v1.1.0

Zod is a TypeScript-first schema library that validates data at runtime and infers static types from the same schema. Use when validating API request bodies, form input, environment variables or config files, parsing untrusted JSON, or migrating from Zod 3 to Zod 4. Covers z.object, safeParse, z.infer, coercion, discriminated unions, refinements, JSON Schema output and the v4 API changes.

93/100 quality
3.87× impact
SAFE· stale score
View skill
$

The Problem

Noa leads product at a 25-person e-commerce company with 15,000 SKUs. Search is a WHERE name ILIKE '%query%' query — it's slow (3 seconds), returns irrelevant results, and doesn't handle typos. Searching "nike shooes" returns nothing. There are no filters (color, size, price range). The "sort by relevance" button sorts alphabetically. 35% of searches return zero results. They're losing $80K/month in revenue from users who search, find nothing, and leave. They need fast, typo-tolerant search with faceted filters, synonym support, and search analytics to know what customers want.

Step 1: Build the Search Engine

typescript
// src/search/engine.ts — Product search with PostgreSQL full-text, facets, and typo tolerance
import { pool } from "../db";
import { Redis } from "ioredis";

const redis = new Redis(process.env.REDIS_URL!);

interface SearchRequest {
  query: string;
  filters?: {
    category?: string[];
    brand?: string[];
    priceMin?: number;
    priceMax?: number;
    color?: string[];
    size?: string[];
    inStock?: boolean;
    rating?: number;           // minimum rating
    tags?: string[];
  };
  sort?: "relevance" | "price_asc" | "price_desc" | "newest" | "popular" | "rating";
  page?: number;
  limit?: number;
  userId?: string;             // for personalization
}

interface SearchResult {
  products: Product[];
  facets: Facets;
  total: number;
  page: number;
  totalPages: number;
  query: string;
  correctedQuery: string | null;  // "Did you mean..."
  suggestions: string[];
  searchId: string;            // for analytics tracking
}

interface Product {
  id: string;
  name: string;
  slug: string;
  description: string;
  price: number;
  originalPrice: number | null;
  image: string;
  brand: string;
  category: string;
  rating: number;
  reviewCount: number;
  inStock: boolean;
  relevanceScore: number;
}

interface Facets {
  categories: Array<{ name: string; count: number }>;
  brands: Array<{ name: string; count: number }>;
  priceRanges: Array<{ label: string; min: number; max: number; count: number }>;
  colors: Array<{ name: string; count: number }>;
  sizes: Array<{ name: string; count: number }>;
  ratings: Array<{ stars: number; count: number }>;
}

// Synonyms map
const SYNONYMS: Record<string, string[]> = {
  "sneakers": ["shoes", "trainers", "kicks"],
  "laptop": ["notebook", "computer"],
  "phone": ["mobile", "smartphone", "cell"],
  "tv": ["television", "monitor", "screen"],
  "hoodie": ["sweatshirt", "pullover"],
  "pants": ["trousers", "jeans"],
  "tee": ["t-shirt", "tshirt"],
  "couch": ["sofa", "loveseat"],
};

// Main search function
export async function search(req: SearchRequest): Promise<SearchResult> {
  const query = req.query.trim();
  const page = req.page || 1;
  const limit = req.limit || 24;
  const offset = (page - 1) * limit;
  const searchId = `s-${Date.now().toString(36)}`;

  // Typo correction
  const correctedQuery = await correctTypos(query);
  const searchQuery = correctedQuery || query;

  // Expand with synonyms
  const expandedTerms = expandSynonyms(searchQuery);

  // Build PostgreSQL full-text search query
  const tsQuery = expandedTerms
    .map((term) => term.split(/\s+/).map((w) => `${w}:*`).join(" & "))
    .join(" | ");

  // Build WHERE conditions
  const conditions: string[] = [];
  const params: any[] = [];
  let paramIndex = 1;

  // Full-text search
  conditions.push(`search_vector @@ to_tsquery('english', $${paramIndex})`);
  params.push(tsQuery);
  paramIndex++;

  // Filters
  if (req.filters?.category?.length) {
    conditions.push(`category = ANY($${paramIndex})`);
    params.push(req.filters.category);
    paramIndex++;
  }
  if (req.filters?.brand?.length) {
    conditions.push(`brand = ANY($${paramIndex})`);
    params.push(req.filters.brand);
    paramIndex++;
  }
  if (req.filters?.priceMin !== undefined) {
    conditions.push(`price >= $${paramIndex}`);
    params.push(req.filters.priceMin);
    paramIndex++;
  }
  if (req.filters?.priceMax !== undefined) {
    conditions.push(`price <= $${paramIndex}`);
    params.push(req.filters.priceMax);
    paramIndex++;
  }
  if (req.filters?.color?.length) {
    conditions.push(`colors && $${paramIndex}`);
    params.push(req.filters.color);
    paramIndex++;
  }
  if (req.filters?.size?.length) {
    conditions.push(`sizes && $${paramIndex}`);
    params.push(req.filters.size);
    paramIndex++;
  }
  if (req.filters?.inStock) {
    conditions.push("stock_count > 0");
  }
  if (req.filters?.rating) {
    conditions.push(`avg_rating >= $${paramIndex}`);
    params.push(req.filters.rating);
    paramIndex++;
  }

  const whereClause = conditions.join(" AND ");

  // Sort
  let orderBy: string;
  switch (req.sort || "relevance") {
    case "relevance":
      orderBy = `ts_rank_cd(search_vector, to_tsquery('english', $1)) * (1 + ln(1 + sales_count)) DESC`;
      break;
    case "price_asc": orderBy = "price ASC"; break;
    case "price_desc": orderBy = "price DESC"; break;
    case "newest": orderBy = "created_at DESC"; break;
    case "popular": orderBy = "sales_count DESC"; break;
    case "rating": orderBy = "avg_rating DESC, review_count DESC"; break;
    default: orderBy = "ts_rank_cd(search_vector, to_tsquery('english', $1)) DESC";
  }

  // Execute search
  const [productsResult, countResult, facetsResult] = await Promise.all([
    pool.query(
      `SELECT id, name, slug, description, price, original_price, image_url, brand, category,
              avg_rating, review_count, stock_count > 0 as in_stock,
              ts_rank_cd(search_vector, to_tsquery('english', $1)) as relevance
       FROM products WHERE ${whereClause}
       ORDER BY ${orderBy}
       LIMIT $${paramIndex} OFFSET $${paramIndex + 1}`,
      [...params, limit, offset]
    ),
    pool.query(`SELECT COUNT(*) FROM products WHERE ${whereClause}`, params),
    getFacets(whereClause, params),
  ]);

  const total = parseInt(countResult.rows[0].count);
  const products = productsResult.rows.map((r: any) => ({
    id: r.id, name: r.name, slug: r.slug,
    description: r.description?.slice(0, 200),
    price: r.price, originalPrice: r.original_price,
    image: r.image_url, brand: r.brand, category: r.category,
    rating: parseFloat(r.avg_rating || "0"), reviewCount: r.review_count,
    inStock: r.in_stock, relevanceScore: parseFloat(r.relevance || "0"),
  }));

  // Suggestions for zero results
  const suggestions = total === 0 ? await getSuggestions(query) : [];

  // Track search analytics
  trackSearch(searchId, query, total, req.filters, req.userId).catch(() => {});

  return {
    products, facets: facetsResult, total, page,
    totalPages: Math.ceil(total / limit),
    query, correctedQuery, suggestions, searchId,
  };
}

async function getFacets(whereClause: string, params: any[]): Promise<Facets> {
  const [categories, brands, colors, sizes, ratings] = await Promise.all([
    pool.query(`SELECT category as name, COUNT(*) as count FROM products WHERE ${whereClause} GROUP BY category ORDER BY count DESC LIMIT 20`, params),
    pool.query(`SELECT brand as name, COUNT(*) as count FROM products WHERE ${whereClause} GROUP BY brand ORDER BY count DESC LIMIT 20`, params),
    pool.query(`SELECT UNNEST(colors) as name, COUNT(*) as count FROM products WHERE ${whereClause} GROUP BY name ORDER BY count DESC`, params),
    pool.query(`SELECT UNNEST(sizes) as name, COUNT(*) as count FROM products WHERE ${whereClause} GROUP BY name ORDER BY count DESC`, params),
    pool.query(`SELECT FLOOR(avg_rating) as stars, COUNT(*) as count FROM products WHERE ${whereClause} GROUP BY stars ORDER BY stars DESC`, params),
  ]);

  return {
    categories: categories.rows,
    brands: brands.rows,
    priceRanges: [
      { label: "Under $25", min: 0, max: 2500, count: 0 },
      { label: "$25 - $50", min: 2500, max: 5000, count: 0 },
      { label: "$50 - $100", min: 5000, max: 10000, count: 0 },
      { label: "$100 - $200", min: 10000, max: 20000, count: 0 },
      { label: "Over $200", min: 20000, max: 999999, count: 0 },
    ],
    colors: colors.rows,
    sizes: sizes.rows,
    ratings: ratings.rows.map((r: any) => ({ stars: parseInt(r.stars), count: parseInt(r.count) })),
  };
}

async function correctTypos(query: string): Promise<string | null> {
  // Check if query matches products well
  const { rows } = await pool.query(
    `SELECT word FROM ts_stat('SELECT search_vector FROM products')
     WHERE word % $1 AND word != $1 ORDER BY similarity(word, $1) DESC LIMIT 1`,
    [query.toLowerCase()]
  );
  return rows[0]?.word || null;
}

function expandSynonyms(query: string): string[] {
  const words = query.toLowerCase().split(/\s+/);
  const expanded = [query];

  for (const word of words) {
    for (const [key, synonyms] of Object.entries(SYNONYMS)) {
      if (word === key || synonyms.includes(word)) {
        expanded.push(query.replace(new RegExp(word, "gi"), key));
        for (const syn of synonyms) {
          if (syn !== word) expanded.push(query.replace(new RegExp(word, "gi"), syn));
        }
      }
    }
  }

  return [...new Set(expanded)];
}

async function getSuggestions(query: string): Promise<string[]> {
  const popular = await redis.zrevrange("search:popular", 0, 9);
  return popular.filter((s) => s.toLowerCase().includes(query.toLowerCase().slice(0, 3))).slice(0, 5);
}

async function trackSearch(searchId: string, query: string, results: number, filters: any, userId?: string): Promise<void> {
  await redis.zincrby("search:popular", 1, query.toLowerCase());
  if (results === 0) await redis.zincrby("search:zero_results", 1, query.toLowerCase());

  await pool.query(
    `INSERT INTO search_analytics (search_id, query, result_count, filters, user_id, created_at)
     VALUES ($1, $2, $3, $4, $5, NOW())`,
    [searchId, query, results, JSON.stringify(filters || {}), userId]
  );
}

Results

  • Zero-result searches: 35% → 5% — typo correction handles "nike shooes"; synonym expansion matches "sneakers" when searching "trainers"
  • Search latency: 3s → 120ms — PostgreSQL GIN index on tsvector column; facet queries run in parallel
  • Revenue recovered: $80K/month — users find what they want; search-to-purchase conversion rate up 3x
  • Faceted filters — color, size, price range, brand, rating filters narrow 15,000 products to exactly what the customer wants
  • Search analytics reveal demand — "wireless earbuds" is the top zero-result query; merchandising team adds the category; 200 sales in the first week

Related use cases