tavily

PackSearch

Tavily is a plugin that gives an AI agent access to real-time web data through search, extract, crawl, and research APIs. It lets the agent search the live web, pull content from specific pages, and run multi-step online research as part of its work.

Unavailable. Delivery for this kind is on the roadmap — not serving yet.

Have a Tavily account and API credentials available.

What your AI can do with it

  • Search the live web for current information
  • Extract content from specific web pages
  • Crawl websites to gather data across multiple pages
  • Run multi-step online research tasks
  • Build AI applications on top of Tavily's APIs

Getting started

  1. Have a Tavily account and API credentials available.
  2. Add the Tavily plugin to your agent setup.
  3. Configure the plugin with your API credentials.
  4. Ask your agent to search the web, extract page content, or run research to confirm it works.

Signals

GitHub stars
37k
Forks
4k
Last commit
Sep 2026

Questions

What can the agent do with Tavily?
It can search the live web, extract content from pages, crawl websites, and run multi-step online research using Tavily's search, extract, crawl, and research APIs.
Do I need an API key?
The plugin connects to Tavily's APIs, so you need Tavily credentials to configure it. Check Tavily's documentation for account setup details.
Is this a plugin or an MCP server?
It is a plugin in the search category.
Advanced
Item type
plugin
Key
anthropics-claude-plugins-official-tavily
Source
github.com/anthropics/claude-plugins-official