An asynchronous document parsing skill powered by Sugon-Scnet for structured layout analysis, markdown conversion, and high-precision OCR of public files.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install document-parse-ocr
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 document-parse-ocr 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 Sugon-Scnet Document Intelligent OCR?
The Sugon-Scnet Document Intelligent OCR skill is an advanced asynchronous document processing agent designed to seamlessly parse high-volume files. By integrating this capability into your AI workflows via Openclaw Skills, you can submit any publicly accessible file URL (PDF, image, etc.) and automatically extract rich structural metadata. The skill accurately recognizes paragraphs, headers, structured tables, charts, complex mathematical formulas, footnotes, headers, footers, and official stamps.
This skill is built for developers who require high-throughput document intelligence without blocking active processes. Instead of waiting on synchronous API responses, it dispatches tasks asynchronously, polls for status updates, and returns a temporary download URL containing the structured JSON results and formatted Markdown. Utilizing this capability within Openclaw Skills ensures robust document-to-data conversion pipelines for your AI agents.
Sugon-Scnet Document Intelligent OCR Use Cases
- High-Volume Contract Analysis: Automate the parsing of lengthy PDF agreements to extract tabular lease terms, payment schedules, and stamp locations.
- Academic & Technical Research: Extract text, embedded charts, and complex formulas from scientific papers or manuals for search indexation.
- Financial Reporting Workflows: Transform multi-page financial statements into structured JSON objects to analyze balance sheets and income tables dynamically.
- Markdown Content Generation: Automatically convert scanned PDFs and document layouts into clean, formatted Markdown for downstream LLM ingestion.
How Sugon-Scnet Document Intelligent OCR Works
- Job Submission: The user or AI agent triggers the skill with a publicly reachable document URL (
fileUrl) and the desired recognition type (ocrType, defaulting toDOC_PARSING). - Asynchronous Dispatch: The skill sends an asynchronous processing request to the Sugon-Scnet OCR API, initiating a non-blocking background task.
- Polling and Status Checks: The skill's internal script automatically polls the Scnet API, monitoring job progress up to a configurable timeout (default 600 seconds) with handling for potential rate limits (HTTP 429) and retry mechanisms.
- Structured JSON & Markdown Assembly: Once complete, the Scnet API yields a structured JSON output (mapping tables, text boxes, and stamps) alongside a fully formatted Markdown translation of the document.
- Temporary Artifact Delivery: The skill outputs the parsed structured content directly, providing a download link for the output files that remains valid for 12 hours.
Sugon-Scnet Document Intelligent OCR Setup
Prerequisites
- Python 3.6+
- A Sugon-Scnet API Token. Register at the Scnet Website and obtain your key from the console (format:
sc-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx).
Step 1: Install Dependencies
Run the following command in your terminal to install the necessary library:
pip install requests
Step 2: Configure Environment Variables
Create a .env file inside the config/ directory of the skill with your credentials:
# ===== Sugon-Scnet OCR API Configuration =====
SCNET_API_KEY=your_scnet_api_key_here
SCNET_API_BASE=https://api.scnet.cn/api/llm/v1
Ensure that you set proper file permissions on the .env file (e.g., chmod 600 config/.env) to secure your API key.
Sugon-Scnet Document Intelligent OCR Data Schema & Taxonomy
Input Parameters
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
ocrType |
String | No | DOC_PARSING |
The document analysis type to perform. Currently only supports DOC_PARSING. |
fileUrl |
String | Yes | - | Publicly accessible HTTP/HTTPS URL of the target document file. |
Output Schema
Upon successful task completion, the skill returns a structured JSON payload containing the following hierarchy:
- Metadata: Information about the file (page count, execution status).
- Structural Elements: An array of identified elements such as:
paragraphs: Plain text blocks mapping directly to reading order.tables: Extracted table content serialized into arrays or markdown formats.charts&stamps: Positions and visual classifications of stamps and illustrations.markdown: A consolidated markdown representation of the entire document.
- Error Log: In case of failure, detailed fields including
error_codeanderror_messageare logged.
Sugon-Scnet Document Intelligent OCR Advanced Features
- Asynchronous Long-Running Task Support: Includes a built-in polling client capable of waiting up to 10 minutes (configurable via
POLL_TIMEOUT), designed specifically to handle large files without causing AI agent connection dropouts. - Automatic Rate-Limit Handling: Integrates intelligent backing-off and automatic retries (up to 3 times) on encountering HTTP
429 Too Many Requestsresponses. - Multi-Format Semantic Parsing: Leverages advanced OCR techniques to identify layout blocks, keeping tables, formulas, and headers contextually isolated rather than returning scrambled text blocks.
- AI Agent Native Prompt Integration: Seamlessly hooks into agent ecosystems as part of Openclaw Skills, allowing conversational triggers (e.g., "Analyze the contract at this link") to execute full-scale data extraction pipelines.
SKILL.md
Loading
Related Openclaw Skills
Bank Draft OCR
An AI-powered OCR tool to extract core metadata such as issue dates, payees, and transaction amounts from bank drafts and financial bills.
Sugon-Scnet Telegraphic Transfer Voucher OCR
A highly specialized financial OCR tool for automatically parsing and extracting core transaction fields from telegraphic transfer bank vouchers.
Payment Voucher OCR
An OCR skill designed to recognize and accurately extract key financial information from payment and withdrawal vouchers.
skill-review-pro
An advanced AI skill quality assurance and static evaluation system designed to score, review, and harden agentic instructions using modular, domain-specific policies.
ADP Global Invoice Extraction Free
A zero-registration, high-precision AI document processing tool designed to extract structured JSON data from global invoices and receipts in over 100 languages.
Seal OCR (Sugon-Scnet)
A specialized OCR tool for recognizing, localizing, and extracting text and metadata from official seals on documents and invoices.
Featured*
