A high-precision OCR skill designed to detect and recognize text in complex layouts, multi-language documents, and single-page PDFs.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install tencentcloud-ocr-generalaccurate
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 tencentcloud-ocr-generalaccurate 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 Tencent Cloud General Accurate OCR?
The Tencent Cloud General Accurate OCR skill provides developers with a robust solution for extracting text from visual media where precision is paramount. This skill is particularly effective for processing images with high text density, complex structural layouts, or multi-language content including Chinese, English, digits, and special characters. As part of the Openclaw Skills ecosystem, it allows for seamless integration of cloud-based optical character recognition into automated workflows.
Beyond simple text extraction, this skill supports single-page PDF recognition and can provide granular word-level information, such as character coordinates and confidence scores. This makes it an essential tool for developers building applications that require deep analysis of documents like legal files, shop signs, or street-view imagery where standard OCR might fail.
Tencent Cloud General Accurate OCR Use Cases
- Digitizing complex legal documents and contracts with high accuracy requirements.
- Extracting text from street-view images or shop signs for location-based data processing.
- Processing multi-language resumes to feed into structured data pipelines.
- Converting single-page PDF scans into searchable and editable text formats.
- Performing high-precision data extraction from cluttered or low-quality network images.
How Tencent Cloud General Accurate OCR Works
- The user provides an image via a URL or a Base64 encoded string.
- The skill initializes the Tencent Cloud SDK using provided environment credentials.
- A request is sent to the GeneralAccurateOCR endpoint, optionally specifying PDF processing or word-level detail.
- Tencent Cloud's high-precision models analyze the image to detect text blocks and recognize characters.
- The skill receives the raw text and metadata, returning a clean JSON object to the Openclaw Skills caller.
Tencent Cloud General Accurate OCR Setup
To use this skill, ensure you have Python 3.6+ installed and the necessary SDK:
pip install tencentcloud-sdk-python
Configure your Tencent Cloud credentials as environment variables:
export TENCENTCLOUD_SECRET_ID="your_secret_id"
export TENCENTCLOUD_SECRET_KEY="your_secret_key"
Run the recognition script using a remote image URL:
python scripts/main.py --image-url "https://example.com/document.jpg"
Tencent Cloud General Accurate OCR Data Schema & Taxonomy
The skill accepts several configuration parameters and returns a structured JSON response:
| Parameter | Type | Required | Description |
|---|---|---|---|
| ImageUrl | string | Optional* | The URL of the image to process (preferred). |
| ImageBase64 | string | Optional* | Base64 encoded image data (max 10MB). |
| IsPdf | boolean | No | Enable PDF recognition (default: false). |
| IsWords | boolean | No | Return word-level information (default: false). |
Output Format:
{
"raw_text": "Extracted text content...",
"RequestId": "unique-request-id-from-tencent"
}
Tencent Cloud General Accurate OCR Advanced Features
- PDF Support: Specifically handles single-page PDF documents for text extraction.
- Word-Level Metadata: Optional retrieval of coordinates and confidence levels for every recognized word.
- Resume Parsing: Specialized guidance for structured information extraction from multi-language resumes.
- Source Tracking: Integrated UserAgent tracking to identify calls originating from Openclaw Skills for better logging and analytics.
SKILL.md
Loading
Related Openclaw Skills
Model Catalog Updater
A specialized tool for Openclaw Skills that automates the discovery and integration of LLM endpoints into your local configuration.
Infoxmed VIP Membership QR Generator
An automated utility for generating Infoxmed VIP membership activation QR codes through natural language processing.
iPlay Media Controller
A streamlined automation skill that allows AI agents to trigger media playback in the iPlay desktop application using any supported URL.
MakeAIClips — AI Video Clipper
MakeAIClips is a powerful AI-driven tool that automatically converts long-form YouTube videos into high-engagement vertical clips with burned-in captions and viral hook titles.
dev-toolkit
A comprehensive suite of seven developer utility APIs for hashing, conversion, math, DNS, and text transformation.
Temporary Email Manager
An automated utility to create, manage, and monitor temporary email addresses through a secure REST API integration.
Featured*
