| title: ISP Handbook OCR Engine | |
| emoji: 📄 | |
| colorFrom: blue | |
| colorTo: indigo | |
| sdk: docker | |
| pinned: false | |
| license: mit | |
| app_port: 7860 | |
| # ISP Handbook OCR Engine | |
| Extracts structured content from uploaded handbook PDFs using a hybrid | |
| text + OCR pipeline. Supports table detection, real-time editing, and | |
| multi-format export (PDF, DOCX, HTML, JSON). | |
| ## Endpoints | |
| | Method | Path | Description | | |
| |--------|----------------------|------------------------------------| | |
| | GET | `/` | Health probe | | |
| | GET | `/health` | Detailed health check | | |
| | GET | `/docs` | Swagger UI | | |
| | POST | `/extract` | Plain text extraction | | |
| | POST | `/extract-structured`| Structured extraction with tables | | |
| | POST | `/export` | Export edited content to file | | |
| | POST | `/save` | Persist to platform database | | |