Nykaa

Fetch full product details for a single Nykaa product: ingredients, how-to-use instructions, review summary, available variants/shades, offers, images, and stock status. Call this after `search_products` when the user asks for deeper information about a specific product. Always use a `product_id` returned by the most recent `search_products` result — never guess or invent an ID.

Search the Nykaa beauty catalog (makeup, skincare, haircare, fragrance, bath & body). Call this whenever the user wants to find, discover, compare, or shortlist beauty products. Do not use it for fashion products like apparel, footwear, handbags, wallets, luggage, jewellery, sunglasses, or other fashion accessories. QUERY NORMALIZATION RULE: Convert the user's request into a clean, minimal query using only the core keywords explicitly mentioned by the user. Remove filler words (e.g., "show me", "I want", "can you find") and keep only the essential product terms. Preserve the original intent and wording as much as possible. Keep gender, skin type, hair type, finish, formulation, SPF level, conscious/ethical attributes (vegan, cruelty-free, clean), and preference/claims (paraben-free, fragrance-free, etc.) directly in the `query` text — the search engine handles these well as keywords. Do NOT extract them into separate parameters. Extract a term into a dedicated filter only for: price ("under ₹500"), discount, or rating. When extracted into a filter, omit only that exact term from 'query.' If removing extracted filter terms would make `query` empty, keep the most important user-provided product/category term in `query`. Do not add inferred filters, sort orders, synonyms, or extra keywords. The default sort is relevance — always use relevance unless the user explicitly requests a different sort order. Examples: - "matte lipstick for oily skin" → query: "matte lipstick oily skin" - "vegan gel cleanser SPF 50" → query: "vegan gel cleanser SPF 50" - "paraben-free shampoo for curly hair" → query: "paraben-free shampoo curly hair" - "glossy lipstick for women under 500" → query: "glossy lipstick for women", price_max: 500 For vague recommendation queries, use the closest searchable terms already present in the user's wording; add new keywords only when the query would otherwise be empty or impossible to search. If the user query contains a lot of constraints and number of results are less than 10, prompt the user with alternate queries for viewing more products. For "show more" / "next page" requests, increment page and keep all other parameters unchanged. The tool returns products with product names,price,discount,ratings and reviews.After results return summarise the results with the available product data in a table,DO NOT specify what data is missing. DO NOT recommend only 1 product as the best among the all. DO NOT specify which product is not an exact match to the query