HN Showcase
Back to browse
GitHub Repository

An Unofficial MCP server that lets Claude Desktop search products, manage carts, and complete Blinkit orders automatically using natural language commands.

33 starsPython

Blinkit MCP – Let Claude order groceries

by hereisSwapnil·Mar 5, 2026·4 points·0 comments
Visit ProjectView on HN

AI Analysis

●MidRabbit HoleShip It

Browser automation MCP for one Indian grocer; interesting proof-of-concept, zero production use case.

Strengths
  • •Playwright browser control cleanly decouples agent from Blinkit API surface—conceptually generalizable to other sites lacking APIs or with restrictive ToS.
  • •Complete end-to-end flow (login via OTP, search, cart, checkout) shipped and working today with one-click installation via .mcpb bundle.
Weaknesses
  • •Entirely dependent on Blinkit website structure; single CSS selector change breaks all automation—unmaintainable fragility versus official APIs or partner integrations.
  • •Single-website proof-of-concept with no reusable abstraction, pattern library, or claim to generality; MCP ecosystem already has dozens of single-service connectors without this specific value.
Target Audience

Claude Desktop users, MCP developers experimenting with browser-based agent automation

Similar To

Anthropic MCP ecosystem · Playwright · Selenium automation scripts

Post Description

I built a small MCP server that lets Claude Desktop order groceries from Blinkit using natural language.

Example prompt:

“Buy milk from Blinkit”

Claude then:

• searches products • adds them to the cart • selects the delivery address • proceeds to checkout

Under the hood the flow looks like this:

Claude Desktop → MCP tool call → Python MCP server → Playwright browser automation → Blinkit website

Instead of relying on APIs, the agent interacts with Blinkit by controlling a real browser session through Playwright.

So Claude effectively becomes an agent that can complete a real world shopping task. The project started as an experiment around a simple question: If LLMs can control browsers reliably, do we still need service APIs for automation? Curious what other MCP integrations people are experimenting with.

Similar Projects

Developer Tools●●Solid

Open-source MCP server that lets AI assistants shop via Google's UCP

Exposes the entire UCP shopping flow as discrete MCP tools (ucp_discover, checkout create/update/fulfillment/complete), which is smart because agents need typed, deterministic primitives not brittle browser hacks. It’s practical — pip installable with a CLI entrypoint and basic tests — so you can prototype agent-driven purchases fast. What would push it higher: richer merchant adapter examples, more end-to-end payment/edge-case docs, and clearer security/payment handling guidance.

Niche GemShip It
nguthrie
204mo ago
AI/ML●●●Banger

MCP plugin that lets Claude autonomously pay for APIs via Lightning

Agents finally have bank accounts: 402 interception + Lightning payments just works.

Zero to OneBold BetBig Brain
jaafarskafi
102mo ago