Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →PaddleOCR document parsing skill based on PaddleOCR-VL-1.5. Provides SOTA-level document understanding with ultra-high precision recognition and parsing. Use when user needs to parse, extract, or understand document content.
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -57% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -31% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -51% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -52% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -58% | 0% |
基于 SOTA 文档解析模型 PaddleOCR-VL-1.5 构建，为 Agent 加上"眼睛"，对文档进行超高精度识别、解析。
export BAIDU_API_KEY="your_key"
百度文档/表格 OCR 识别，需设置 BAIDU_OCR_AK 和 BAIDU_OCR_SK。
bashpython3 scripts/baidu_ocr_doc.py doc /path/to/document.jpg python3 scripts/baidu_ocr_doc.py table /path/to/table.png
Other measured skills in the registry, with their headline benchmark lift.