Web Extractor for Openclaw

A high-performance web content extraction skill that uses Jina AI to convert noisy URLs into clean, LLM-ready Markdown text.

kukuxnd
v1.0.0
Mar 21, 2026
1
1.1k
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install web-extractor

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 web-extractor 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 Extractor?

Web Extractor is a specialized tool designed for the Openclaw Skills ecosystem to streamline the way AI agents interact with web content. By leveraging the r.jina.ai engine, it strips away non-essential elements like scripts, navigation bars, and advertisements, leaving only the core substantive text.

This process transforms cluttered HTML into streamlined Markdown, significantly reducing token consumption and improving the accuracy of AI analysis. It is an essential utility for developers building research-oriented agents or automated news aggregators within the Openclaw Skills framework.

Web Extractor Use Cases

  • Summarizing long-form news articles or technical documentation from a URL.
  • Extracting core content from technical blogs for knowledge base injection.
  • Reducing token usage by filtering out HTML noise before sending data to an LLM.
  • Automated research gathering where clean text is required from multiple web sources.

How Web Extractor Works

  1. The skill receives a target URL via a trigger word such as extract web content or summarize news.
  2. It makes an automated request to the Jina AI reader API to process the URL.
  3. The raw HTML is converted into clean, readable Markdown text.
  4. The resulting content is saved to a local directory (e.g., /tmp/) as a .md file.
  5. The AI agent reads the local file to perform analysis, summarization, or data extraction.

Web Extractor Setup

To integrate this into your environment, ensure you have curl installed. No complex API keys are required for basic usage with Openclaw Skills.

# Example of manual extraction to test the setup
curl -s https://r.jina.ai/<URL> > /tmp/web-content.md

Web Extractor Data Schema & Taxonomy

The skill follows a simple file-based data organization to ensure compatibility across different Openclaw Skills workflows.

Feature Specification
Output Format Markdown (.md)
Default Storage /tmp/ directory
Data Cleaning Automatic removal of CSS, JS, and Nav elements
Input Standard HTTP/HTTPS URLs

Web Extractor Advanced Features

  • Support for custom output paths to integrate with persistent storage or database ingestion.
  • Zero-load AI processing by pre-filtering content to maximize context window efficiency.
  • Multi-agent support where one agent extracts and another performs complex reasoning on the resulting file.
  • Compatible with any news site, technical blog, or article page globally.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*