Tavily MCP

Overview

Tavily is an AI-powered search API designed specifically for AI agents, providing real-time web search, content extraction, and website crawling with high accuracy and low latency. With the Tavily MCP integration in SquadOS, your agents can search for up-to-date information on the web, extract raw content from URLs, map website structures, and conduct in-depth research on any topic autonomously.

• Official website: https://tavily.com/
• Composio documentation: docs.composio.dev/toolkits/tavily_mcp

Authentication

This tool uses OAuth 2.0 with dynamic client registration (DCR_OAUTH). In the managed SquadOS/Composio flow, authorize your Tavily account on the hosted page; do not paste an API key into prompts, messages, or knowledge bases.

How to connect in SquadOS

1. Go to Tools in the side menu (/admin/tools).
2. Open the Available tab and search for Tavily MCP.
3. Open the Configure tool panel and click Activate. The integration moves to Active with the Awaiting authentication state.
4. Open the active tool and click Connect now.
5. Authorize the Tavily account on the secure page hosted by Composio and complete OAuth.
6. Back in SquadOS, confirm the account under Accounts and use Edit actions to allow only the operations you need.
7. In the agent editor, open Tools, click Add Tool, choose Tavily MCP, select the account and actions, save, and run a controlled test. (See Organization Tools for the complete flow.)

Available actions

The product’s local export, checked on June 17, 2026, contains five Tavily MCP actions; all are documented below. Frontmatter records schema snapshot 20260303_09. Check the fields shown in the SquadOS editor before saving because provider parameters and limits can evolve.

Use TAVILY_MCP_TAVILY_SEARCH for a focused search and TAVILY_MCP_TAVILY_RESEARCH when the task requires synthesis across multiple sources. TAVILY_MCP_TAVILY_EXTRACT reads known URLs; TAVILY_MCP_TAVILY_MAP discovers structure; and TAVILY_MCP_TAVILY_CRAWL traverses and extracts pages.

Caution

External content is untrusted input: it may be wrong or contain malicious instructions. Tell the agent to treat page content as a source, not as a command; restrict domains and paths where possible; preserve source URLs; and verify important claims. Limit depth, breadth, and result count to control latency and usage. Respect authentication, robots.txt, site terms, copyright, and personal data.

Tavily crawl

TAVILY_MCP_TAVILY_CRAWL

Crawl a website starting from a URL. Extracts content from pages with configurable depth and breadth.

Input parameters

Name	Type	Required	Description
url	string	Yes	The root URL to begin the crawl.
limit	integer	No	Total number of links the crawler will process before stopping.
format	string ("markdown" | "text")	No	The format of the extracted web page content. markdown returns content in markdown format. text returns plain text and may increase latency.
max_depth	integer	No	Max depth of the crawl. Defines how far from the base URL the crawler can explore.
max_breadth	integer	No	Max number of links to follow per level of the tree (i.e., per page).
instructions	string	No	Natural language instructions for the crawler. Instructions specify which types of pages the crawler should return.
select_paths	array	No	Regex patterns to select only URLs with specific path patterns (e.g., /docs/.*, /api/v1.*).
extract_depth	string ("basic" | "advanced")	No	Advanced extraction retrieves more data, including tables and embedded content, with higher success but may increase latency.
allow_external	boolean	No	Whether to return external links in the final response.
select_domains	array	No	Regex patterns to restrict crawling to specific domains or subdomains (e.g., ^docs\.example\.com$).
include_favicon	boolean	No	Whether to include the favicon URL for each result.

Output

Name	Type	Required	Description
data	string	Yes	Data from the action execution.
error	string	No	Error message if execution failed.
successful	boolean	Yes	Whether the action executed successfully.

Tavily extract

TAVILY_MCP_TAVILY_EXTRACT

Extract content from URLs. Returns raw page content in markdown or text format.

Input parameters

Name	Type	Required	Description
urls	array	Yes	List of URLs to extract content from.
query	string	No	Query to rerank content chunks by relevance.
format	string ("markdown" | "text")	No	Output format.
extract_depth	string ("basic" | "advanced")	No	Controls extraction level. Check the current schema to balance broader coverage against latency; this parameter does not bypass site authentication or protection.
include_images	boolean	No	Include images from pages.
include_favicon	boolean	No	Include favicon URLs.

Output

Name	Type	Required	Description
data	string	Yes	Data from the action execution.
error	string	No	Error message if execution failed.
successful	boolean	Yes	Whether the action executed successfully.

Tavily map

TAVILY_MCP_TAVILY_MAP

Map a website’s structure. Returns a list of URLs found starting from the base URL.

Input parameters

Name	Type	Required	Description
url	string	Yes	The root URL to begin the mapping.
limit	integer	No	Total number of links the crawler will process before stopping.
max_depth	integer	No	Max depth of the mapping. Defines how far from the base URL the crawler can explore.
max_breadth	integer	No	Max number of links to follow per level of the tree (i.e., per page).
instructions	string	No	Natural language instructions for the crawler.
select_paths	array	No	Regex patterns to select only URLs with specific path patterns (e.g., /docs/.*, /api/v1.*).
allow_external	boolean	No	Whether to return external links in the final response.
select_domains	array	No	Regex patterns to restrict crawling to specific domains or subdomains (e.g., ^docs\.example\.com$).

Output

Name	Type	Required	Description
data	string	Yes	Data from the action execution.
error	string	No	Error message if execution failed.
successful	boolean	Yes	Whether the action executed successfully.

Tavily research

TAVILY_MCP_TAVILY_RESEARCH

Perform comprehensive research on a given topic or question. Use this tool when you need to gather information from multiple sources, including web pages, documents, and other resources, to answer a question or complete a task. Returns a detailed response based on the research findings.

Input parameters

Name	Type	Required	Description
input	string	Yes	A comprehensive description of the research task.
model	string ("mini" | "pro" | "auto")	No	Defines the degree of depth of the research. mini is good for narrow tasks with few subtopics. pro is good for broad tasks with many subtopics.

Output

Name	Type	Required	Description
data	string	Yes	Data from the action execution.
error	string	No	Error message if execution failed.
successful	boolean	Yes	Whether the action executed successfully.

Tavily search

TAVILY_MCP_TAVILY_SEARCH

Search the web for current information on any topic. Use for news, facts, or data beyond your knowledge cutoff. Returns snippets and source URLs.

Input parameters

Name	Type	Required	Description
query	string	Yes	Search query.
topic	string	No	The category of the search. This will determine which of our agents will be used for the search.
country	string	No	Boost search results from a specific country. This will prioritize content from the selected country in the search results. Available only if topic is general.
end_date	string	No	Will return all results before the specified end date. Required to be written in the format YYYY-MM-DD.
start_date	string	No	Will return all results after the specified start date. Required to be written in the format YYYY-MM-DD.
time_range	string	No	The time range back from the current date to include in the search results.
max_results	integer	No	The maximum number of search results to return.
search_depth	string ("basic" | "advanced" | "fast" | "ultra-fast")	No	The depth of the search. basic for generic results, advanced for more thorough search, fast for optimized low latency with high relevance, ultra-fast for prioritizing latency above all else.
include_images	boolean	No	Include a list of query-related images in the response.
exclude_domains	array	No	List of domains to specifically exclude, if the user asks to exclude a domain set this to the domain of the site.
include_domains	array	No	A list of domains to specifically include in the search results, if the user asks to search on specific sites set this to the domain of the site.
include_favicon	boolean	No	Whether to include the favicon URL for each result.
include_raw_content	boolean	No	Include the cleaned and parsed HTML content of each search result.
include_image_descriptions	boolean	No	Include a list of query-related images and their descriptions in the response.

Output

Name	Type	Required	Description
data	string	Yes	Data from the action execution.
error	string	No	Error message if execution failed.
successful	boolean	Yes	Whether the action executed successfully.