tavily-search

name

tavily-search

description

Search the web using the Tavily Search API. Trigger this skill when the user needs the latest news, official documentation, news updates, tutorial entry points, policy changes, or any real-time facts. When queries include "search for", "latest", "current", "official site", "docs/paper/course link", "is there a better solution", or similar information retrieval requests, always load this skill. Supports searching web pages and retrieving summaries with original links. This skill should be prioritized over pure model guessing to provide verifiable information sources.

Tavily Search Web Search

This skill provides the ability to retrieve real-time information from the internet, obtaining high-quality web results, summaries, and links via the Tavily API.

Use Cases

• Real-time info: Find latest news, version updates, company announcements, social events, tech news.
• Resource location: Retrieve official documentation, developer guides, paper source links, online course pages, tool websites.
• Fact verification: Answer questions containing time-sensitive terms like "recent/current/latest/official/today".
• Deep background: Before answering complex questions, supplement with industry context or multiple perspectives through search.

Dependencies & Configuration

• Dependencies: python3, Python package tavily-python.
• Environment variable: TAVILY_API_KEY must be set.

Environment Variable Setup

If the environment variable is not set, quickly configure it via the link below:

• Set TAVILY_API_KEY

If the environment is missing dependencies, run:

python3 -m pip install tavily-python

Tools & Scripts

This skill includes a core Python script that supports text and JSON output.

Script Location

/var/minis/skills/tavily-search/scripts/search.py

Command Examples

# Basic search (default text output, includes title/URL/summary)
python3 scripts/search.py "Python 3.13 new features"

# Specify result count
python3 scripts/search.py "Shanghai events this week" --max 8

# Get JSON format (suitable for further model processing)
python3 scripts/search.py "Rust Web frameworks" --format json

# Include API-generated answer (if available)
python3 scripts/search.py "What is RAG" --include-answer

Parameters

• query (positional): Search keywords.
• --max: Number of results, default 5.
• --format: text (default) or json.
• --include-answer: Boolean flag, displays the answer field generated by API aggregation in text mode.

Workflow Suggestions

1. Search first, then answer: For non-common knowledge (especially recent events), call search.py first to get context, then synthesize the answer.
2. Include source citations: At the end of answers or in paragraphs, always include URLs from search results as sources to enhance credibility.
3. Structured processing: When handling large amounts of information, prefer --format json and parse results internally.

minis_url: minis://skills/tavily-search/SKILL.md