A specialized web search skill designed for AI agents to discover URLs, conduct technical research, and generate cited summaries from structured web results.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install tavily-tool
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 tavily-tool 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 Web Search & Discovery?
Tavily is a search engine optimized for AI agents and LLMs, providing clean, high-utility data for research tasks. By integrating this capability through Openclaw Skills, developers can bypass the noise of traditional search engines and access structured information that is easy for agents to parse and analyze. It is particularly effective for gathering up-to-date links, performing deep lead research, and producing cited summaries from diverse web results.
This skill provides a bridge between your automated workflows and the live web, ensuring your agents have access to the most current information available. Whether you are building a research bot or an automated lead generation tool, this addition to your library of Openclaw Skills significantly enhances the agent's ability to verify facts and find authoritative documentation.
Tavily Web Search & Discovery Use Cases
- Rapidly discover URLs and high-quality sources for specific technical queries.
- Conduct deep lead research and gather up-to-date information for competitive analysis.
- Generate cited summaries by extracting data from multiple relevant web results.
- Filter search results to include or exclude specific domains for targeted data collection.
How Tavily Web Search & Discovery Works
- The user or agent provides a search query and optional parameters like result count or domain filters to the skill.
- The skill authenticates with the Tavily API using a secure environment variable.
- Search results are retrieved and formatted, with structured JSON sent to stdout for programmatic processing.
- A simplified URL list is simultaneously sent to stderr for quick manual verification or logging.
- The agent consumes this data to synthesize answers, extract links, or trigger subsequent Openclaw Skills in the workflow.
Tavily Web Search & Discovery Setup
To get started with Tavily within the Openclaw Skills framework, you must have a valid API key. Set it in your environment and run the search script as follows:
export TAVILY_API_KEY="your_api_key_here"
node skills/tavily/scripts/tavily_search.js --query "best rust http client" --max_results 5
You can also use the included shell wrapper for easier CLI access: scripts/tavily_search.sh.
Tavily Web Search & Discovery Data Schema & Taxonomy
The skill processes search parameters and returns structured data as follows:
| Parameter | Type | Description |
|---|---|---|
| query | String | The search string used to find information on the web. |
| max_results | Number | The maximum number of search results to return (default is 5). |
| include_domains | String | A comma-separated list of specific domains to include in results. |
| exclude_domains | String | A comma-separated list of domains to filter out of results. |
| urls-only | Boolean | If set, the script returns only the list of discovered URLs. |
Tavily Web Search & Discovery Advanced Features
- Domain-level filtering to restrict searches to authoritative sources like official documentation or academic sites.
- Multi-output streams where JSON data is prioritized for automation while human-readable lists are separated.
- Lightweight CLI implementation that can be easily integrated into larger Openclaw Skills pipelines.
- Support for specific search fields including query optimization for LLM-based discovery.
SKILL.md
Loading
Related Openclaw Skills
Agent Docs
A framework for creating documentation specifically optimized for LLM consumption, RAG retrieval, and token efficiency.
Agent Backlink Network (ABN)
A decentralized protocol for AI agents to discover, negotiate, and execute backlink trades or purchases without intermediaries.
Better Notion
A comprehensive Notion API integration providing full CRUD capabilities for pages, databases, and content blocks.
Apple Photos Integration
A specialized tool for searching, listing, and exporting media from the macOS Photos.app library using script-driven SQLite queries.
X-Read
A browser-based automation tool for extracting and summarizing public X (Twitter) content into clean markdown without requiring an account.
bitclawden
A professional-grade interface for managing Bitwarden credentials and secrets directly through the Bitwarden CLI.
Featured*
