About Tavily MCP Server
Tavily MCP Server exposes Tavily search and URL extraction through a FastAPI-based remote SSE server. It is for MCP clients that need configurable web search and content retrieval, with API-key authentication for the server endpoints and Tavily API.
Key features
- 01Tavily search with depth and topic options
- 02Search filters for time and domains
- 03Optional images and raw content
- 04Extract content from URLs
Use cases
- 01Search the web for general or news topics
- 02Limit search results by time range or domain
- 03Retrieve content from selected URLs
MCP support
Uses fastapi-mcp to expose its Tavily API endpoints as a remote SSE MCP server for MCP clients.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMETavily MCP Server
Basic FastAPI-based server application that provides endpoints for interacting with the Tavily API. The server includes API key authentication and uses the fastapi-mcp library to create a remote SSE MCP server out of it.
Installation
To install the project's dependencies, run the following command:
pip install -r requirements.txt
Usage
To run the server, use the following command:
uvicorn server:app --host 0.0.0.0 --port 8000 --reload
The server will be available at http://127.0.0.1:8000.
API Key
The server requires two API keys:
API_KEY: The API key for accessing the server's endpoints.TAVILY_API_KEY: The API key for accessing the Tavily API.
Set these API keys as environment variables before running the server.
Endpoints
/tavily-search
Perform a search using the Tavily API.
- Method: POST
- Request Body:
TavilySearchRequestquery(str): The search query.search_depth(Literal["basic", "advanced"], optional): The depth of the search. Defaults to "basic".topic(Literal["general", "news"], optional): The topic of the search. Defaults to "general".days(int, optional): The number of days to search. Defaults to 3.time_range(Literal["day", "week", "month", "year", "d", "w", "m", "y"], optional): The time range of the search. Defaults to "day".max_results(int, optional): The maximum number of results to return. Defaults to 10.include_images(bool, optional): Whether to include images in the results. Defaults to False.include_image_descriptions(bool, optional): Whether to include image descriptions in the results. Defaults to False.include_raw_content(bool, optional): Whether to include raw content in the results. Defaults to False.include_domains(list[str], optional): A list of domains to include in the search. Defaults to an empty list.exclude_domains(list[str], optional): A list of domains to exclude from the search. Defaults to an empty list.
- Response:
TavilyResponse
/tavily-extract
Extract content from URLs using the Tavily API.
- Method: POST
- Request Body:
TavilyExtractRequesturls(list[str]): A list of URLs to extract content from.extract_depth(Literal["basic", "advanced"], optional): The depth of the extraction. Defaults to "basic".include_images(bool, optional): Whether to include images in the extraction. Defaults to False.
- Response:
TavilyResponse
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.