A multi-service content extraction tool that converts URLs, articles, and complex file formats into clean, structured text.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install universal-extractor
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 universal-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 Universal Extractor?
The Universal Extractor is a versatile utility within the ecosystem of Openclaw Skills designed to solve the challenge of web and document noise. It provides four specialized micro-services that allow AI agents and developers to strip away advertisements, navigation menus, and footers from web pages, or convert binary files like PDFs and DOCX into machine-readable text.
By leveraging this skill, users can ensure their LLMs receive only the most relevant context, reducing token waste and improving the accuracy of downstream tasks. Whether you are building a research bot or an automated data pipeline, this tool simplifies the ingestion of unstructured data from the web and local storage.
Universal Extractor Use Cases
- Cleaning web articles for LLM consumption by removing ads and HTML clutter.
- Generating instant summaries of long-form online content using integrated LLM features.
- Converting legacy document formats like PDF and DOCX into structured JSON or plain text.
- Automating data extraction from various file types uploaded via base64 encoding.
How Universal Extractor Works
- The user or agent sends a POST request to one of the four specialized endpoints based on the source material.
- For URLs, the service fetches the remote content and applies cleaning algorithms to isolate the primary text.
- For documents, the service decodes the base64 payload and parses the internal structure of the file (PDF, DOCX, etc.).
- Optional summarization logic is applied if the article extraction endpoint is used with the summarize flag enabled.
- The system returns a structured JSON response containing the extracted text, metadata like word count, and author information.
- Payments are handled seamlessly via the x402 protocol on the Base network, eliminating the need for traditional API keys.
Universal Extractor Setup
To integrate this component of Openclaw Skills into your workflow, ensure you have a wallet capable of sending USDC on the Base network for the x402 protocol payments. No complex API registration is required.
# Example: Extracting clean text from a URL using cURL
curl -X POST https://api.openclaw.io/x402s/clean-url \
-H "Content-Type: application/json" \
-d '{"url": "https://example.com/article"}'
Universal Extractor Data Schema & Taxonomy
The Universal Extractor organizes output into structured JSON objects. Below is the metadata taxonomy for the extraction services:
| Field | Type | Description |
|---|---|---|
text |
String | The primary cleaned content or extracted document body. |
title |
String | The identified title of the article or document. |
author |
String | The extracted author name, if available. |
word_count |
Integer | Total number of words in the extracted text. |
summary |
String | (Optional) LLM-generated summary of the content. |
date |
String | Publication or creation date of the source material. |
Universal Extractor Advanced Features
- Multi-format support including PDF, DOCX, TXT, and CSV via base64 ingestion.
- Integrated LLM summarization on the /extract-article endpoint for immediate insights.
- Protocol-level payment integration using USDC on Base, enabling frictionless agent-to-agent transactions.
- High-speed cleaning algorithms that remove boilerplate content while preserving semantic structure.
SKILL.md
Loading
Related Openclaw Skills
Sexy-Claw
Sexy-Claw is an automated AI agent skill that discovers and recommends high-value social media influencers and video content based on personalized aesthetic preferences.
AI Reliability Layer
A collection of three micro-services focused on repairing broken JSON, validating response rules, and generating guaranteed structured data for AI agents.
Data Model Selector
A technical decision framework for selecting between relational, document, and graph data models based on data shape and access patterns.
Encoding Format Advisor
A technical decision-making tool for selecting data encoding formats and designing schema evolution strategies that maintain system compatibility.
Expense Assistant
An intelligent reimbursement management assistant that automates receipt recognition, travel allowance calculation, and organized document filing.
Stock Pulse
An intelligent A-share decision dashboard that transforms raw stock data into actionable AI-powered buy, sell, and hold signals.
ideas2tasks
An automated system that converts raw ideas into structured agile tasks using a virtual Scrum team.
Featured*
