tavily-skill_skill

Overview

This skill integrates the Tavily API to perform real-time web search and extract page content for research, news, and current-information tasks. It requires a valid Tavily API key and supports configurable search depth, result limits, and optional content extraction. Use it when you need up-to-date web results or summarized answers drawn from live sources. The skill is optimized for quick setup and flexible query parameters.

How this skill works

The skill sends search requests to Tavily with your query, search_depth (basic or advanced), and max_results, then returns structured results and optional raw page content or AI-generated answers. It can include image URLs and full page content when requested, and respects the API key and rate limits configured in your environment. Error checks include API key presence, connection tests, and handling of Tavily rate-limit responses.

When to use it

• You need real-time web results or current news on a topic
• Researching a recent event, trend, or development
• Aggregating multiple web sources into a concise answer
• Extracting page content or images from live web pages
• When you have a Tavily API key and want automated search workflows

Best practices

• Store the Tavily API key in environment variables or your app config for security
• Start with search_depth=basic and increase to advanced only when you need deeper context
• Limit max_results to 5–10 to balance coverage and response size
• Request include_raw_content only when you need full page text to avoid large payloads
• Handle rate-limit responses gracefully and back off per Tavily plan limits

Example use cases

• Quickly gather the latest technology or policy news and return a short summary with source links
• Perform background research on a topic (e.g., market trends) using advanced searches and include an AI-generated answer
• Extract article content or images from multiple pages for downstream analysis or summarization
• Integrate into a chatbot or workflow to provide current web-backed answers during conversations
• Automate monitoring queries that fetch recent mentions or developments on a subject