A specialized web search and content extraction skill optimized specifically for AI agents to retrieve clean, high-relevance data.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install test-web-searcher
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install test-web-searcher 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 Search?
Tavily Search is a professional-grade tool integrated into the Openclaw Skills library, designed to solve the problem of noisy search results in AI workflows. By leveraging the Tavily API, this skill bypasses the clutter of traditional search engines, returning structured and concise snippets that are ready for immediate ingestion by Large Language Models.
This skill is particularly valuable for developers building autonomous agents that need to perform real-time research or verify information. It provides a bridge between the static knowledge of a model and the dynamic, ever-changing landscape of the live web, ensuring that your AI applications remain accurate and well-informed.
Tavily Search Use Cases
- Performing real-time factual verification for AI-generated content.
- Conducting deep-dive research on complex technical topics using advanced search modes.
- Monitoring current events and news cycles for specific industries or keywords.
- Extracting clean, readable text from specific URLs to populate RAG pipelines.
- Automating daily information gathering tasks for competitive intelligence.
How Tavily Search Works
- The user or autonomous agent initiates a query through the CLI scripts provided in the skill directory.
- The skill processes the input parameters, such as search depth and topic filters.
- A request is sent to the Tavily API, which utilizes AI-optimized indexing to find the most relevant web content.
- The raw data is parsed into clean text snippets, removing ads, navigation bars, and irrelevant HTML elements.
- The refined results are returned to the agent in a format that maximizes token efficiency for LLM context windows.
Tavily Search Setup
To integrate this search capability into your Openclaw Skills, you must first obtain an API key from Tavily and ensure Node.js is installed.
# Set your environment variable
export TAVILY_API_KEY="your_tavily_api_key"
# Run a basic search
node scripts/search.mjs "latest trends in AI agents"
# Run a deep research search
node scripts/search.mjs "quantum computing breakthroughs 2024" --deep
Tavily Search Data Schema & Taxonomy
The skill manages search results and extraction data through a structured command-line interface. Below are the primary data configurations available:
| Feature | Parameter | Description |
|---|---|---|
| Result Count | -n |
Determines the number of relevant snippets returned (max 20). |
| Search Mode | --deep |
Triggers a more thorough, multi-pass search for complex queries. |
| Topic Focus | --topic |
Switches between 'general' and 'news' specialized indexes. |
| Time Window | --days |
Filters news results to a specific number of recent days. |
| Extraction | extract.mjs |
Specifically targets a URL to pull core body content. |
Tavily Search Advanced Features
- Deep Search optimization for multi-source synthesis and complex problem solving.
- News-specific filtering to ensure temporal relevance for trending topics.
- Specialized content extraction script to transform cluttered URLs into LLM-friendly text.
- Seamless compatibility with other Openclaw Skills for building multi-step automation workflows.
- Configurable result density to balance between detailed information and token conservation.
SKILL.md
Loading
Related Openclaw Skills
Tavily Search
A high-performance web search engine specifically optimized to provide AI agents with clean, context-rich information.
Toggl CLI
A feature-rich command-line interface for interacting with Toggl Track workspaces, enabling seamless time management directly from the terminal.
Specter CLI
A powerful command-line interface for the Specter platform to automate company enrichment and professional data management.
Notion CLI
A production-grade command-line interface for the Notion API, enabling seamless workspace management through the terminal.
security-guard
A specialized security enforcement skill designed to protect API keys, credentials, and personal data across all AI agent interactions.
YouYou
A natural language health assistant for managing vitals, medical reports, and multi-specialty health analysis locally.
METADATA
Featured*
