Aggregates multiple web search engines with intelligent caching, cost-aware routing, and result fusion for AI agents.
Sponsored
Designed for AI agents, this Rust-based MCP server streamlines web search by consolidating results from Brave, Exa, SearXNG, Tavily, and DuckDuckGo into a single, efficient tool. It intelligently manages search costs and performance through aggressive caching, cost-aware routing to the cheapest available backend, and sophisticated result fusion to deduplicate and enhance findings across multiple engines. This eliminates the complexity and expense of juggling multiple search APIs, providing a unified, optimized solution for AI-native web search.
Características Principales
01Aggregates 5 popular search backends (Brave, Exa, SearXNG, Tavily, DuckDuckGo)
02Implements cost-aware routing to optimize query expenses
03Features a two-tier caching system for high hit rates
04Performs intelligent result fusion with deduplication and score normalization
05Supports batch searching and advanced query filters (category, time range)
060 GitHub stars
Casos de Uso
01Reducing API costs and complexity when integrating multiple search engines
02Enabling developers to build AI applications with reliable, cached search capabilities
03Providing efficient and cost-effective web search for AI agents
