Tavily Search V2

Web search using Tavily's LLM-optimized API. Returns relevant results with content snippets, scores, and metadata.

Security Scan

This skill is internally consistent with its stated purpose: it requires Node and a Tavily API key and the included script simply calls Tavily's API and formats results; no unrelated credentials, installs, or surprising behavior were found (aside from minor metadata mismatches).

Purpose & Capability

Name and description (web search via Tavily) match the requirements and code: only Node and TAVILY_API_KEY are required, and the script POSTs to https://api.tavily.com/search. Nothing in the files demands unrelated services or permissions.

Instruction Scope

SKILL.md instructions stay on-topic: they show how to provide an API key (env or config) and how to run the provided Node script with search options. The runtime instructions do not ask the agent to read unrelated files, system secrets, or exfiltrate data to unexpected endpoints.

Install Mechanism

There is no install spec (instruction-only plus an included script), so nothing is downloaded or installed automatically. The included script is small and plain JS; it makes a network call to the declared Tavily API host only.

Credentials

Only a single environment variable (TAVILY_API_KEY) is required and is justified by the API call. The SKILL.md suggests storing the key in OpenClaw config or as an env var — standard for API clients.

Persistence & Privilege

The skill is not always-enabled and is user-invocable; it does not request elevated system persistence or modify other skills' configurations. Autonomous invocation is allowed (platform default) but not combined with other concerning permissions.

Assessment

This skill looks coherent and only needs your Tavily API key and Node. Before installing: verify you trust the tavily.com service and the skill publisher; prefer setting TAVILY_API_KEY as an environment variable rather than pasting it into shared config files; review and rotate the key if you later delete the skill; note minor metadata mismatches in the bundled _meta.json (slug/version/owner differ from registry) — these are likely benign but you may want to confirm the publisher identity if provenance matters. Also ensure your Node version supports global fetch or run in an environment where fetch is available.

Tavily Search

Search the web and get relevant results optimized for LLM consumption.

Authentication

Get your API key at https://tavily.com and add to your OpenClaw config:

{
  "skills": {
    "entries": {
      "tavily-search": {
        "enabled": true,
        "apiKey": "tvly-YOUR_API_KEY_HERE"
      }
    }
  }
}

Or set the environment variable:

export TAVILY_API_KEY="tvly-YOUR_API_KEY_HERE"

Quick Start

Using the Script

node {baseDir}/scripts/search.mjs "query"
node {baseDir}/scripts/search.mjs "query" -n 10
node {baseDir}/scripts/search.mjs "query" --deep
node {baseDir}/scripts/search.mjs "query" --topic news

Examples

# Basic search
node {baseDir}/scripts/search.mjs "python async patterns"

# With more results
node {baseDir}/scripts/search.mjs "React hooks tutorial" -n 10

# Advanced search
node {baseDir}/scripts/search.mjs "machine learning" --deep

# News search
node {baseDir}/scripts/search.mjs "AI news" --topic news

# Domain-filtered search
node {baseDir}/scripts/search.mjs "Python docs" --include-domains docs.python.org

Options

Option	Description	Default
-n <count>	Number of results (1-20)	10
--depth <mode>	Search depth: ultra-fast, fast, basic, advanced	basic
--topic <topic>	Topic: general or news	general
--time-range <range>	Time range: day, week, month, year	-
--include-domains <domains>	Comma-separated domains to include	-
--exclude-domains <domains>	Comma-separated domains to exclude	-
--raw-content	Include full page content	false
--json	Output raw JSON	false

Search Depth

Depth	Latency	Relevance	Use Case
ultra-fast	Lowest	Lower	Real-time chat, autocomplete
fast	Low	Good	Need chunks but latency matters
basic	Medium	High	General-purpose, balanced
advanced	Higher	Highest	Precision matters, research

Tips

• Keep queries under 400 characters - Think search query, not prompt
• Break complex queries into sub-queries - Better results than one massive query
• Use --include-domains to focus on trusted sources
• Use --time-range for recent information
• Filter by score (0-1) to get highest relevance results