SERP Maven

•
124
•

Scrape Google search results using a real Chrome browser and residential proxies to deliver structured SERP data for applications and AI agents.

SERP Maven is a self-hosted alternative to commercial search APIs that extracts Google search results using real Chrome browser automation paired with residential proxies. By leveraging intelligent session warming, hybrid HTTP fetching between anti-bot challenges, and built-in protection against poisoned results, it reliably outputs structured JSON containing organic results, ads, related queries, local map packs, and AI Overviews. It can be run as a CLI tool, embedded directly into Python workflows, deployed as an HTTP API, or exposed to AI agents via an MCP server.

Características Principales

01124 GitHub stars
02Hybrid mode combining full browser automation with direct HTTPS fetching for reduced bandwidth and latency
03Session warming and poisoned results protection to maintain trusted identities and bypass anti-scraping blocks
04Comprehensive SERP parsing for organic results, ads, related questions, local map packs, and AI Overviews
05Native Model Context Protocol (MCP) server integration to give AI agents live web search capabilities
06Multiple operational modes including Python library, CLI, local HTTP API, and pre-built Docker container

Casos de Uso

01Monitor competitor rankings and track SEO keyword positions without relying on costly third-party SERP APIs
02Harvest search engine data and query trends at scale using custom residential proxy networks
03Equip AI agents and LLM applications with live Google search and real-time SERP context via MCP