Web Search Skill for Openclaw

A comprehensive web search and content extraction skill that enables AI agents to retrieve real-time data using Jina AI and DuckDuckGo without mandatory API keys.

flashery
v1.0.1
Feb 21, 2026
0
0
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install yves-web-search

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 yves-web-search 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 Web Search Skill?

The web-search skill is a robust tool designed to provide AI agents with real-time internet access and high-quality content extraction capabilities. By utilizing a multi-method approach, it allows agents within Openclaw Skills to perform searches and fetch full-page content in Markdown format, which is ideal for LLM processing. The skill prioritizes accessibility by offering free tiers and no-config options, making it an essential component for any research-oriented agent.

As a core part of the Openclaw Skills ecosystem, this tool overcomes the limitations of traditional search APIs. It leverages Jina AI Reader for clean content extraction and DuckDuckGo Lite for lightweight, captcha-free searching. Whether you are building a news aggregator or a technical documentation assistant, this skill provides the reliability and flexibility needed for modern AI workflows.

Web Search Skill Use Cases

  • Finding and fetching the latest software documentation or API references.
  • Researching current events and news to ground AI responses in real-time data.
  • Automating competitive analysis by extracting structured information from competitor websites.
  • Fact-checking claims by searching across multiple regional search engines.

How Web Search Skill Works

  1. The agent receives a search query and uses DuckDuckGo Lite to identify high-relevance URLs.
  2. The agent selects the best target URL from the search results based on the user's intent.
  3. The selected URL is processed through the Jina AI Reader API to convert HTML content into clean Markdown.
  4. If the primary web_fetch method encounters blocks, the skill triggers a Python-based fallback using the ddgs library to ensure data retrieval.

Web Search Skill Setup

Basic usage within Openclaw Skills requires no configuration. Simply use the provided fetch commands. For Python-based fallbacks or advanced features, follow these steps:

# Install the Python fallback library
pip install ddgs

# (Optional) Set up Jina API key for advanced extraction
export JINA_API_KEY="your_key_here"

You can then run searches directly via terminal commands or within your agent's environment using the web_fetch tool.

Web Search Skill Data Schema & Taxonomy

The skill organizes its operations around simple URL parameters and script flags to interface with Openclaw Skills effectively.

Component Method Data Format
Search DuckDuckGo Lite URL Query Params
Extraction Jina AI Reader Markdown String
Advanced Jina API Script JSON Output
Fallback Python ddgs List of Objects

Web Search Skill Advanced Features

  • Support for regional search filters (e.g., us-en, de-de) to localize results.
  • Advanced Jina Reader modes including 'search' and 'ground' for verified information retrieval.
  • Custom CSS selectors and element removal to clean up fetched web content.
  • Headless browser integration for navigating complex, JavaScript-heavy websites when necessary.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*