mcp-websearch

mcp-websearch

0
research_and_datacloud_platforms

This project is a Model Context Protocol (MCP) server designed to perform web searches across Indonesian news portals and Wikipedia. It offers integration with large language models, flexible LLM provider options, and secure data practices. The application operates as a Google Cloud Function with Python 3.11.

Web Search MCP Server

A Model Context Protocol (MCP) server for searching content from Indonesian news portals and Wikipedia. It can be deployed as a Google Cloud Function.

Key Benefits

  • Pre-built integrations for LLMs
  • Flexibility with LLM providers
  • Secure data practices

Architecture Overview

  • MCP Hosts: Access data via MCP
  • MCP Clients: 1:1 server connections
  • MCP Servers: Expose capabilities
  • Data Sources: Local and remote

System Requirements

  • Python 3.11 or higher
  • Google Cloud SDK
  • Serper API key

Local Development Setup

  1. Clone the repository
  2. Create a virtual environment
  3. Install dependencies
  4. Set API key
  5. Run locally

Deploying to Google Cloud Functions

  1. Install and initialize Google Cloud SDK
  2. Set API key
  3. Run deployment script

Function Usage

Send HTTP POST requests with JSON body format for search queries. Available portals include detik, liputan6, cnn, and wikipedia.

Technical Details

  • Runtime: Python 3.11
  • Async implementation
  • Memory: 256MB
  • Timeout: 60 seconds

Error Handling

Returns HTTP status codes: 200, 400, 404, 500 with MCP format responses.