aresbit/claude-skills/paddleocr-doc-parsing
Claude Code and agent skill by aresbit/claude-skills
- community
Checking that the SKILL.md file is still in the repository.
Document parsing and OCR using PaddleOCR. Extract text, tables, and structured data from PDFs and images (JPG, PNG, BMP, TIFF). Supports 110+ languages and outputs Markdown format preserving document layout. Use when processing documents for data extraction, invoice parsing, form recognition, or converting scanned documents to structured text. Requires PADDLEOCR_API_URL and PADDLEOCR_ACCESS_TOKEN environment variables.
Install
npx vskill@latest install aresbit/claude-skills/paddleocr-doc-parsingSKILL.md
The same SKILL.md was published earlier as aresbit/claude-skills/paddleocr-doc-parsing.
PaddleOCR Document Parsing
This skill integrates Baidu's PaddleOCR to provide powerful document parsing capabilities for Claude Code.
Features
- Multi-format support: PDF, JPG, PNG, BMP, TIFF
- Layout analysis: Automatically detects text, tables, formulas, headers, and other document elements
- Multi-language OCR: Covers 110+ languages including Chinese, English, Japanese, Korean
- Structured output: Returns Markdown format that preserves document hierarchy
- Direct PDF support: No conversion needed, PDFs parsed directly via API
Setup
1. Get API Credentials
Visit PaddleOCR to get API credentials:
- Register and verify your phone number
- Select "PaddleOCR-VL-1.5" model
- Copy TOKEN and API_URL from the example code
- Free daily quota: tens of thousands of pages
2. Configure Environment Variables
Add to ~/.zshrc or ~/.claude/.env:
Usage
Parse PDF Document
Output:
- Markdown printed to stdout
{filename}_ocr/directory created in same folder as PDF- Each page saved as separate
.mdfile - Images extracted and saved to output directory
Options:
Parse Image
Same script handles images (fileType detected automatically):
Scripts
| Script | Purpose | |
|---|---|---|
scripts/parse_pdf_direct.py | Main parsing script for PDF and images | |
scripts/parse_document.py | Alternative script (legacy API format) | |
scripts/extract_fields.py | Field extraction for common document types |
References
references/api_reference.md- API endpoint specificationsreferences/document_types.md- Supported documents and best practices
API Response Format
Error Handling
The API may occasionally return 500 errors due to service load. The script includes automatic retry logic. If parsing fails:
- Wait a few seconds
- Retry the request
- Check API credentials are correct
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingpaddleocr-doc-parsing/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/aresbit/claude-skills/paddleocr-doc-parsing)