{
  "markdown": "# silicon-PaddleOCR\n\nA Claude Code plugin that provides OCR (Optical Character Recognition) capabilities using PaddleOCR via the SiliconFlow API.\n\n## Features\n\n- Extract text from images (JPG, PNG, WebP, BMP, GIF)\n- Batch processing with glob patterns\n- JSON output format for programmatic use\n- Customizable recognition prompts\n- Support for custom models and parameters\n\n## Installation\n\n1. Clone this repository to your local Claude Code plugins directory:\n```bash\ngit clone https://github.com/aotenjou/silicon-PaddleOCR.git ~/.claude/plugins/silicon-PaddleOCR\n```\n\n2. Set up your SiliconFlow API key as an environment variable:\n```bash\nexport SILICONFLOW_API_KEY=\"your_api_key_here\"\n```\n\n3. Install required Python dependencies:\n```bash\npip install openai\n```\n\n## Usage\n\n### Via Claude Code Command\n\nAfter installing the plugin, use the `/ocr` command in Claude Code:\n\n```\n/ocr /path/to/image.jpg\n```\n\n### Direct Script Usage\n\nYou can also run the OCR script directly:\n\n```bash\n# Single image\npython3 skills/ocr/scripts/ocr_skill.py /path/to/image.jpg\n\n# Multiple images with glob pattern\npython3 skills/ocr/scripts/ocr_skill.py /path/to/images/*.png\n\n# JSON output\npython3 skills/ocr/scripts/ocr_skill.py --json /path/to/image.jpg\n\n# Custom prompt\npython3 skills/ocr/scripts/ocr_skill.py -p \"Extract as Markdown table\" /path/to/table.jpg\n\n# Save results to file\npython3 skills/ocr/scripts/ocr_skill.py --json --output results.json /path/to/images/*.jpg\n```\n\n## Script Arguments\n\n| Argument | Description |\n|----------|-------------|\n| `images` | Image file path(s) or glob pattern (required) |\n| `-k, --api-key` | API key (default: SILICONFLOW_API_KEY env) |\n| `-m, --model` | OCR model (default: PaddlePaddle/PaddleOCR-VL-1.5) |\n| `-p, --prompt` | Custom recognition prompt |\n| `-j, --json` | Output in JSON format |\n| `-o, --output` | Save results to file |\n| `--max-tokens` | Max tokens in response (default: 300) |\n\n## Configuration\n\nGet your API key from [SiliconFlow](https://siliconflow.cn).\n\n## Supported Image Formats\n\n- JPG/JPEG\n- PNG\n- WebP\n- BMP\n- GIF\n\n## License\n\nMIT License - see [LICENSE](LICENSE) file for details.\n\n## Contributing\n\nContributions are welcome! Please feel free to submit issues or pull requests.\n\n## Project Structure\n\n```\nsilicon-PaddleOCR/\n├── .claude-plugin/\n│   └── plugin.json              # Plugin manifest\n├── commands/\n│   └── ocr.md                   # /ocr command definition\n└── skills/\n    └── ocr/\n        ├── SKILL.md             # Skill documentation\n        ├── scripts/\n        │   └── ocr_skill.py     # Main implementation\n        ├── references/\n        │   └── api-configuration.md\n        └── examples/\n            └── sample-usage.sh\n```\n",
  "bytes": 2699,
  "sha": "91488d1daf0933fa8dce45ce9b4387ae58d82a01448591678789ba8b3632e32f",
  "repo_slug": "aotenjou/silicon-paddleocr",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/skl_aotenjou_silicon_paddleocr_silicon_paddl_2cf693ac/readme"
}