kshern/mcp-tavily
FreeTavily AI 搜索 API (社区实现版本)。
FreeFree tier
Outputs: text
About kshern/mcp-tavily
MCP Tavily is a Model Context Protocol server that wraps the Tavily API, providing advanced search and content extraction capabilities for AI applications. It offers three specialized search tools (basic search, context-aware search, and Q&A search) as well as a content extraction tool. Users can customize search depth, topic, time range, domain filtering, and more via rich configuration options. The server is designed for seamless integration into MCP-compatible environments like Claude Desktop, and can be installed via npx or Smithery.
Key Features
Three search tools: basic search (search), context-aware search (searchContext), and Q&A search (searchQNA)
Content extraction tool (extract) to retrieve content from URLs
Configurable search depth: basic or advanced
Search topic categories: general, news, finance
Time range filtering (year, month, week, day)
Option to include images, image descriptions, answers, and raw content in search results
Domain inclusion and exclusion lists
Maximum results and token limits
Proper error handling with descriptive error messages
Installable via npx or Smithery for easy setup
Pros & Cons
Pros
- Offers three distinct search tools tailored for different use cases
- Supports both basic and advanced search depth for flexibility
- Includes content extraction capability alongside search
- Provides extensive configuration options (time range, domain filters, etc.)
- Clear and consistent response format across all tools
- Open source and free to use
Cons
- Requires a valid Tavily API key
- Dependent on Tavily API availability and rate limits
- No built-in caching or rate limiting (must be handled externally)
- Limited documentation beyond the readme
Best For
Integrating Tavily search into AI assistants and chatbots via MCPBuilding context-aware research tools that understand query contextExtracting and summarizing content from multiple web pagesCreating question-answering systems with Tavily Q&A searchAutomating information retrieval for news monitoring and financial data
FAQ
What is MCP Tavily?
MCP Tavily is a Model Context Protocol (MCP) server that provides a wrapper for the Tavily API, enabling advanced search and content extraction through MCP-compatible applications.
How do I install and use MCP Tavily?
You can install it via npx: add the server to your MCP configuration with command 'npx -y @mcptools/mcp-tavily' and set the TAVILY_API_KEY environment variable. Alternatively, install via Smithery using 'npx -y @smithery/cli install @kshern/mcp-tavily'.
What search tools are available?
Three search tools: 'search' (basic search), 'searchContext' (context-aware search), and 'searchQNA' (question-answering search). Additionally, an 'extract' tool is provided for content extraction from URLs.