Skill: PaddleOCR Document Parsing

Owner: Bobholamovic

Summary: Parse documents using PaddleOCR's API.

Tags: latest:1.0.2

Version history:

v1.0.2 | 2026-02-10T08:00:39.427Z | user

• No changes detected in this release.

v1.0.1 | 2026-02-09T13:29:36.477Z | user

• Added a new "Resource Links" section with quick access to the official website, API documentation, and GitHub.
• Updated the setup instructions to clarify algorithm selection (now lists "PP-StructureV3" as an example).
• No functionality changes; documentation improvements only.

v1.0.0 | 2026-02-05T06:46:10.261Z | user

Initial release of paddleocr-doc-parsing.

• Parse images and PDF files using PaddleOCR's API with structured Markdown output.
• Supports 110+ languages and detects layouts including tables and formulas.
• Requires API credentials and common command line utilities (curl, base64, jq).
• Provides example usage for parsing local files and URLs.
• Detailed response structure and setup instructions included.

Archive index:

Archive v1.0.2: 3 files, 3399 bytes

Files: scripts/paddleocr_parse.sh (4923b), SKILL.md (3030b), _meta.json (140b)

File v1.0.2:SKILL.md

name: paddleocr-doc-parsing description: Parse documents using PaddleOCR's API. homepage: https://www.paddleocr.com metadata: { "openclaw": { "emoji": "📄", "os": ["darwin", "linux"], "requires": { "bins": ["curl", "base64", "jq"], "env": ["PADDLEOCR_API_URL", "PADDLEOCR_ACCESS_TOKEN"], }, }, }

PaddleOCR Document Parsing

Parse images and PDF files using PaddleOCR's API. Supports multiple document parsing algorithms with structured output.

Resource Links

| Resource | Link | | --------------------- | ------------------------------------------------------------------------------ | | Official Website | https://www.paddleocr.com | | API Documentation | https://ai.baidu.com/ai-doc/AISTUDIO/Cmkz2m0ma | | GitHub | https://github.com/PaddlePaddle/PaddleOCR |

Key Features

• Multi-format support: PDF and image files (JPG, PNG, BMP, TIFF)
• Layout analysis: Automatic detection of text blocks, tables, formulas
• Multi-language: Support for 110+ languages
• Structured output: Markdown format with preserved document structure

Setup

1. Obtain credentials from the PaddleOCR official website. Click the “API” button, choose the desired algorithm (e.g., PP-StructureV3, PaddleOCR-VL-1.5), and copy the API URL and the access token.
2. Set environment variables:

export PADDLEOCR_API_URL="https://your-endpoint-here"
export PADDLEOCR_ACCESS_TOKEN="your_access_token"

Usage Examples

Run Script

# Parse local image
{baseDir}/paddleocr_parse.sh document.jpg

# Parse local PDF file
{baseDir}/paddleocr_parse.sh -t pdf document.pdf

# Parse document from URL
{baseDir}/paddleocr_parse.sh -t pdf https://example.com/document.pdf

# Output to stdout (default)
{baseDir}/paddleocr_parse.sh document.jpg

# Save output to file
{baseDir}/paddleocr_parse.sh -o result.json document.jpg

Response Structure

{
  "logId": "unique_request_id",
  "errorCode": 0,
  "errorMsg": "Success",
  "result": {
    "layoutParsingResults": [
      {
        "prunedResult": [...],
        "markdown": {
          "text": "# Document Title\n\nParagraph content...",
          "images": {}
        },
        "outputImages": [...],
        "inputImage": "http://input-image"
      }
    ],
    "dataInfo": {...}
  }
}

Important Fields:

• prunedResult - Contains detailed layout element information including positions, categories, etc.
• markdown - Stores the document content converted to Markdown format with preserved structure and formatting.

Quota Information

See official documentation: https://ai.baidu.com/ai-doc/AISTUDIO/Xmjclapam

File v1.0.2:_meta.json

{ "ownerId": "kn77zppfj1a2fc620aygaf9z9980ewfa", "slug": "paddleocr-doc-parsing", "version": "1.0.2", "publishedAt": 1770710439427 }

Archive v1.0.1: 3 files, 3399 bytes

Files: scripts/paddleocr_parse.sh (4923b), SKILL.md (3030b), _meta.json (140b)

File v1.0.1:SKILL.md

name: paddleocr-doc-parsing description: Parse documents using PaddleOCR's API. homepage: https://www.paddleocr.com metadata: { "openclaw": { "emoji": "📄", "os": ["darwin", "linux"], "requires": { "bins": ["curl", "base64", "jq"], "env": ["PADDLEOCR_API_URL", "PADDLEOCR_ACCESS_TOKEN"], }, }, }

PaddleOCR Document Parsing

Parse images and PDF files using PaddleOCR's API. Supports multiple document parsing algorithms with structured output.

Resource Links

| Resource | Link | | --------------------- | ------------------------------------------------------------------------------ | | Official Website | https://www.paddleocr.com | | API Documentation | https://ai.baidu.com/ai-doc/AISTUDIO/Cmkz2m0ma | | GitHub | https://github.com/PaddlePaddle/PaddleOCR |

Key Features

• Multi-format support: PDF and image files (JPG, PNG, BMP, TIFF)
• Layout analysis: Automatic detection of text blocks, tables, formulas
• Multi-language: Support for 110+ languages
• Structured output: Markdown format with preserved document structure

Setup

1. Obtain credentials from the PaddleOCR official website. Click the “API” button, choose the desired algorithm (e.g., PP-StructureV3, PaddleOCR-VL-1.5), and copy the API URL and the access token.
2. Set environment variables:

export PADDLEOCR_API_URL="https://your-endpoint-here"
export PADDLEOCR_ACCESS_TOKEN="your_access_token"

Usage Examples

Run Script

# Parse local image
{baseDir}/paddleocr_parse.sh document.jpg

# Parse local PDF file
{baseDir}/paddleocr_parse.sh -t pdf document.pdf

# Parse document from URL
{baseDir}/paddleocr_parse.sh -t pdf https://example.com/document.pdf

# Output to stdout (default)
{baseDir}/paddleocr_parse.sh document.jpg

# Save output to file
{baseDir}/paddleocr_parse.sh -o result.json document.jpg

Response Structure

{
  "logId": "unique_request_id",
  "errorCode": 0,
  "errorMsg": "Success",
  "result": {
    "layoutParsingResults": [
      {
        "prunedResult": [...],
        "markdown": {
          "text": "# Document Title\n\nParagraph content...",
          "images": {}
        },
        "outputImages": [...],
        "inputImage": "http://input-image"
      }
    ],
    "dataInfo": {...}
  }
}

Important Fields:

• prunedResult - Contains detailed layout element information including positions, categories, etc.
• markdown - Stores the document content converted to Markdown format with preserved structure and formatting.

Quota Information

See official documentation: https://ai.baidu.com/ai-doc/AISTUDIO/Xmjclapam

File v1.0.1:_meta.json

{ "ownerId": "kn77zppfj1a2fc620aygaf9z9980ewfa", "slug": "paddleocr-doc-parsing", "version": "1.0.1", "publishedAt": 1770643776477 }