reader

Basic Information

• Author: cloudsway
• Version: 1.3.0
• Type: tool (Dify plugin)

Description

cloudsway_reader is a Dify plugin wrapper for the Cloudsway Reader API (Content Reader v1.3.0). It provides capabilities to extract content from URLs (HTML or PDF) with extensive control options including crawl modes, OCR settings, and image handling.

Key Features

• Multi-Format Output: Support HTML, TEXT, and MARKDOWN.
• Smart Crawling: fast (static), quality (dynamic rendering), and auto modes.
• PDF Processing: Support parsing PDF text or returning Base64, with optional Enhanced OCR.
• Image Handling: Option to download images as Base64 and control their placement (in-content or separate list).
• Strict Parameter Mapping: Fully exposes all configuration options from the API v1.3.0 specification.

Usage Guide

1. Prerequisites

• Base Path: The API base path (e.g., search-console4c-pre.cloudsway.net).
• Endpoint: The API endpoint path segment.
• Access Key: Your Bearer Token for authorization.

2. Tool Parameters

This plugin exposes all parameters defined in the Cloudsway Reader v1.3.0 documentation:

Parameter	Type	Required	Default	Description
url	String	Yes	-	The target URL to read.
format	Select	No	TEXT	Output format (TEXT, MARKDOWN, HTML). API parameter: formats.
mode	Select	No	fast	Crawl mode: fast (static), quality (dynamic), auto.
timeout	Int	No	30000	Crawl timeout in milliseconds. Controls API connection time.
totalTimeout	Int	No	-	End-to-end request timeout in milliseconds. Interrupts request if exceeded.
imageDownloadEnable	Bool	No	false	Convert images to base64. (Additional costs apply).
imageInContent	Bool	No	true	If true, base64 images are in the body. If false, in image_base64_list.
pdfExtractEnable	Bool	No	false	true: Return parsed PDF text; false: Return PDF base64. (Additional costs apply).
enhancedOcr	Bool	No	false	Enhance PDF extraction accuracy. Effective only if pdfExtractEnable is true.

3. Response Structure

The tool returns the raw JSON response from the API. Common fields include:

• text / markdown / html: The extracted content.
• metadata: { title, description, keywords }.
• logo: Website logo URL.
• site_name: Website name.
• image_list: Array of image URLs.
• image_base64_list: Array of base64 images (if imageInContent is false).
• pdf_pages: Page count (if target is PDF).
• internal_links / external_links: Link analysis results.

Privacy Policy

• Data Collected: This plugin only collects the target URL (url) provided by the user to perform the read operation.
• Data Usage: URLs are processed in real-time by the Cloudsway Reader API to extract content. No data is permanently stored by the plugin.
• Security: Credentials are stored securely. All transmission is encrypted via HTTPS.
• Support: For privacy-related questions, please contact us at support@cloudsway.ai.

License

This project is open-sourced under the MIT License. See LICENSE for details.