groceries-mcp
Verified Safeby o-b-one
Overview
Orchestrates grocery shopping using a Large Language Model (LLM) agent by connecting to various grocery vendor APIs via the Model Context Protocol (MCP).
Installation
docker-compose up shufersal_mcpEnvironment Variables
- MODEL_ID
- VENDOR
- MCP_ENDPOINT
- INFERENCE_ENDPOINT
- VENDOR_API_KEY
- VENDOR_ACCOUNT_ID
- CART_ID
- USERNAME
- PASSWORD
- USER_DATA_PATH
- DEBUG
Security Notes
The Shufersal provider uses Playwright's `page.evaluate` to execute JavaScript for cart operations. While the scripts themselves are hardcoded in the Python service (not LLM-generated), which mitigates direct LLM-based injection, executing code in a browser context always carries some risk. Authentication details (USERNAME, PASSWORD, API keys) are correctly handled via environment variables. Browser session state is persisted to disk (`auth_state.json` in `USER_DATA_PATH`), which is a standard Playwright feature but requires securing the host system's file access.
Similar Servers
AgentUp
A developer-first framework for building, deploying, and managing secure, scalable, and configurable AI agents, supporting various agent types (reactive, iterative) and the Model-Context Protocol (MCP) for seamless interactions.
rohlik-mcp
Enhances LLMs with grocery shopping capabilities across Rohlik Group's online services.
Polymcp
A comprehensive toolkit and agent framework for building Model Context Protocol (MCP) servers and orchestrating them with Large Language Models (LLMs) across Python and TypeScript environments.
zeromcp
A minimal, pure Python Model Context Protocol (MCP) server for exposing tools, resources, and prompts via HTTP/SSE and Stdio transports.