search

Search the web via Tavily's API and return snippets, scores, and metadata.

Updated Feb 26, 2026
One-click install
npx skills add https://github.com/tony2015116/openclaw-backup --skill search-tony2015116
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: search
Source: https://github.com/tony2015116/openclaw-backup/tree/main/skills/tavily/search
Command: npx skills add https://github.com/tony2015116/openclaw-backup --skill search-tony2015116

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires Tavily API, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the challenge of finding relevant web content on any topic, providing fast and accurate search results optimized for AI consumption.

Core Features & Use Cases

  • Web Search: Offers LLM-optimized search capabilities to find content on any topic.
  • Content Snippets: Provides content snippets, scores, and metadata for search results.
  • Use Case: Efficiently find research papers, articles, or tutorials on specific topics without manual search.

Quick Start

Run the search.sh script with your query, like: ./scripts/search.sh '{"query": "AI news"}'.

Frequently Asked Questions about search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I search the web for LLM-optimized content and metadata?

You can search the web for LLM-optimized content using the Tavily API, which retrieves relevant results complete with content snippets, scores, and metadata optimized for AI consumption.

Do I need a Tavily API key to perform web searches?

Yes, you need a Tavily API key for authentication to perform web searches. The Skill supports both OAuth authentication and API key-based authentication to deliver search results.

How do I execute a web search query using the search.sh script?

To execute a web search query, run the search.sh script with a JSON payload containing your query, like: ./scripts/search.sh '{"query": "AI news"}' to retrieve content snippets and metadata.

What is the best way to find research papers and articles for AI consumption?

The best way to find research papers and articles for AI consumption is using an LLM-optimized web search API. It efficiently retrieves relevant content snippets and metadata without requiring manual searches.

Does Tavily API provide content scores and snippets with search results?

Yes, the Tavily API provides content snippets, scores, and metadata alongside highly relevant search results, ensuring the retrieved web content is properly optimized for downstream AI processing.

Are there limitations when using an LLM-optimized web search for content discovery?

Limitations of LLM-optimized web search include dependency on Tavily API availability and authentication requirements. You must manage API keys to ensure continuous access to content discovery and metadata retrieval.