agentleFS
Sign inStart free
OpenMinisOpenMinis/MinisSkills/tavily-search/SKILL.md

tavily-search

Skill584 starsChanged 59 days agoMITView on GitHub

Use the Tavily Search API to perform web searches. When users need to find the latest information, official documentation, news updates, tutorial links, policy changes, or any real-time facts, prioritize triggering this skill. Be sure to load this skill whenever information retrieval requests appear, such as "look this up," "latest," "now," "official website," "literature/paper/course entry point," or "is there a better solution?" It supports searching web pages and retrieving summaries and original links. This skill should take precedence over pure model guessing to provide users with verifiable sources of information.

  • Reads credentials
  • Installs packages

We split this skill into the rules it gives an agent

5 rules. 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 15 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Resource Location: Search for official documentation, developer manuals, entry points to original papers, online course pages, and official …

    line 16
  2. Distinctive:

    In-Depth Background: Before answering complex questions, use search to gather industry background or multiple perspectives.

    line 18
  3. Distinctive:

    If the environment variable is not configured, you can quickly set it using the following link:

    line 27
  4. Distinctive:

    If a dependency is missing from the environment, run:

    line 31
  5. Very specific:

    Structured Processing: If you need to process a large amount of information, prioritize using --format json and parse the results …

    line 72

“Common” is measured: kept by at least 6 in 10 of the 15 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Tavily Search Web Search

This skill provides the ability to search the internet for information in real time, using the Tavily API to obtain high-quality web results, summaries, and links.

Use Cases

  • Real-Time Information: Find the latest news, version updates, company announcements, social events, and technology news.
  • Resource Location: Search for official documentation, developer manuals, entry points to original papers, online course pages, and official tool websites.
  • Fact-Checking: Answer questions containing time-sensitive terms such as "recent," "current," "latest," "official website," or "today."
  • In-Depth Background: Before answering complex questions, use search to gather industry background or multiple perspectives.

Dependencies and Configuration

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

Environment Variable Configuration

If the environment variable is not configured, you can quickly set it using the following link:

  • Set TAVILY_API_KEY

If a dependency is missing from the environment, run:

python3 -m pip install tavily-python

Tools and Scripts

This skill includes one 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, including title/URL/summary)
python3 scripts/search.py "Python 3.13 New Features"

# Specify the number of results to return
python3 scripts/search.py "Shanghai Events This Week" --max 8

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

# Include the reference answer generated by the API (if any)
python3 scripts/search.py "What is RAG Technology" --include-answer

Parameter Description

  • query (positional argument): Search keyword.
  • --max: Number of results to return. Default is 5.
  • --format: text (default) or json.
  • --include-answer: Boolean switch. In text mode, displays the API-aggregated answer field.

Workflow Recommendations

  1. Search First, Then Answer: For topics that are not general common knowledge, especially recent events, first call search.py to obtain context, then synthesize the answer.
  2. Include Source Citations: At the end of the answer or within paragraphs, be sure to include URLs from the search results as sources to improve credibility.
  3. Structured Processing: If you need to process a large amount of information, prioritize using --format json and parse the results internally.

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

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free