Zepto MCP Server

Give your AI assistant live access to Zepto — India's quick-commerce grocery platform. This actor runs as a Model Context Protocol (MCP) server so agents like Claude, ChatGPT, Cursor, and n8n can search products, pull prices and stock, and resolve delivery stores on demand. It also works as a classic scraper when run normally.

Built on Zepto's mobile API (header-only, no login, no unblocking subscription), so keyword search returns the full priced product grid — reliably.

MCP tools

Tool	What it does
search_zepto_products	Search a keyword and return priced product rows for a location (name, brand, MRP, price, discount, stock, image).
get_zepto_product	Full detail for a product-variant id at a delivery location.
zepto_autocomplete	Search-box autocomplete suggestions for a partial query.
resolve_zepto_store	Resolve the serving store (and serviceability) for a lat/lng or city.

Every tool is location-scoped — pass an Indian city name (e.g. Mumbai) or explicit latitude/longitude, since Zepto prices and stock vary by store.

Connect an AI assistant

The actor runs in Standby mode and speaks MCP over Streamable HTTP at:

https://<your-actor-standby-url>/mcp

Add it to an MCP-capable client (Claude Desktop, Cursor, etc.) as a Streamable HTTP server, using your Apify API token as a bearer token. Then ask naturally:

"What's the cheapest milk on Zepto in Mumbai right now?" "Compare Amul butter prices between Delhi and Bengaluru."

Use cases

• Shopping & price assistants — let an agent answer live grocery price/stock questions.
• Research & monitoring agents — automate assortment and discount checks across cities.
• Workflow automation (n8n, Make) — wire Zepto lookups into agentic pipelines.

Also runs as a scraper

Run the actor normally (not standby) with searchQueries + a location to push normalized product rows to a dataset — handy for scheduled jobs and quick tests.

Example (normal run) input

{
    "searchQueries": ["milk", "amul butter"],
    "location": "Mumbai",
    "maxPagesPerQuery": 1,
    "maxItems": 100
}

Output fields

Product rows include name, brand, pack_size, mrp, selling_price, super_saver_price, discount_percent, in_stock, category, subcategory, image, store_id, location_label, and scraped_at. See the dataset schema for the full list.

FAQ

Do I need an API key? You connect with your Apify token as the bearer token for the MCP endpoint. No Zepto login is required.

Which locations work? Any serviceable point in India — a city name or exact coordinates.

How fresh is the data? Every tool call fetches live prices and stock.

Is this legal? You are responsible for your use of the data. Query publicly available information and comply with Zepto's terms and applicable law.