exa-mcp-server and mcp-omnisearch
These are complements that serve different integration patterns: Exa provides a single specialized search and crawling API, while Omnisearch acts as a unified abstraction layer aggregating multiple disparate search and AI tools, allowing users to leverage both together or pick based on whether they prefer depth in one provider versus flexibility across many.
About exa-mcp-server
exa-labs/exa-mcp-server
Exa MCP for web search and web crawling!
Implements an MCP (Model Context Protocol) server that exposes Exa's search APIs—including specialized code search and company research—via HTTP transport to integrate with AI assistants and code editors (Cursor, VS Code, Claude Desktop, etc.). Offers both basic tools (web search, code context, page crawling) and advanced filtering capabilities, with optional Claude Skills for domain-specific workflows like competitor analysis. Deploys as a hosted endpoint or npm package, requiring only an API key for authentication.
About mcp-omnisearch
spences10/mcp-omnisearch
🔍 A Model Context Protocol (MCP) server providing unified access to multiple search engines (Tavily, Brave, Kagi), AI tools (Perplexity, FastGPT), and content processing services (Jina AI, Kagi). Combines search, AI responses, content processing, and enhancement features through a single interface.
# Technical Summary Implements four consolidated MCP tools (web_search, ai_search, github_search, web_extract) with pluggable provider backends, allowing clients to query multiple APIs through a unified interface while gracefully degrading based on available credentials. Supports advanced search operators native to Brave/Kagi, domain filtering via API parameters, and specialized extractors like Firecrawl's interactive scraping and Kagi's multimodal summarization (pages, videos, podcasts). Designed for integration with AI assistants (Claude Desktop, Cline) via environment-variable configuration with zero hard dependencies on any single provider.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work