Tavily Plus for Openclaw

An advanced AI-augmented search skill providing multi-key rotation, sub-query decomposition, and automated research reporting.

fundou1081
v1.0.0
Apr 1, 2026
0
619
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install tavily-plus

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install tavily-plus using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Tavily Plus?

Tavily Plus is a sophisticated search and research engine designed to elevate the capabilities of AI agents. By integrating advanced LLM-driven intent recognition and query decomposition, this skill transforms simple user prompts into comprehensive, multi-layered research operations. It ensures high reliability and throughput for Openclaw Skills by implementing an intelligent API key rotation system that automatically handles rate limits and failovers.

Beyond basic search, Tavily Plus provides deep synthesis of information, allowing agents to provide factual, news-based, or comparative insights with high precision. Whether you are building a research assistant or a real-time news monitor, this skill provides the robust infrastructure needed for high-quality information retrieval and intelligent summarization.

Tavily Plus Use Cases

  • Conducting deep academic or market research using the research mode.
  • Comparing multiple products, technologies, or concepts with the compare intent.
  • Generating offline Markdown or PDF reports for search sessions automatically.
  • Scaling search operations across multiple API keys to avoid rate-limiting issues.
  • Extracting clean content from specific URLs for data processing workflows.

How Tavily Plus Works

  1. The user submits a natural language query which is analyzed by an LLM for intent classification such as factual, news, or comparison.
  2. Complex queries are decomposed into a series of atomic sub-questions to ensure maximum search coverage.
  3. The skill executes these queries via the Tavily API, intelligently rotating through a list of configured API keys if a 429 rate limit error occurs.
  4. Search results from all sub-queries are aggregated and synthesized by an AI model to create a coherent, intelligent summary.
  5. If requested, the system generates a timestamped offline report in the local workspace directory.
  6. The final response, including the summary and metadata, is returned to the agent.

Tavily Plus Setup

To install and configure Tavily Plus for your Openclaw Skills environment, add your API keys to your configuration file or environment variables.

# Example environment variable setup
export TAVILY_API_KEY="your_primary_key"
export TAVILY_API_KEY_2="your_secondary_key"

Configuration via plugins.entries.tavily.config.webSearch:

{
  "apiKey": {
    "key1": "tvly-xxx1",
    "key2": "tvly-xxx2"
  },
  "baseUrl": "https://api.tavily.com"
}

Tavily Plus Data Schema & Taxonomy

Tavily Plus organizes search data and generated artifacts within a structured directory and JSON response format.

Attribute Description
Intent Classified category of the search such as factual or research
Sub-questions Array of strings representing decomposed queries
Doc Path Local path to the generated report: ~/.openclaw/workspace/reports/
Key Rotation Metadata logs indicating which API key was utilized for each request

Generated reports follow a naming convention: tavily-plus-{timestamp}.md.

Tavily Plus Advanced Features

  • Automatic multi-key rotation logic for high-volume enterprise search tasks.
  • Intent-based routing that adjusts search depth and topic focus based on query type.
  • Customizable query modes including research and quick for performance tuning.
  • Automated sub-question decomposition for handling multifaceted or ambiguous requests.
  • Built-in URL extraction capabilities to fetch raw content directly from search results.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Requires
Github Stars: 0
forks: 0

Featured*