# brick-strength-recognition

> 从砖强度检验表格（PDF/图像）中提取结构化字段，对表格类型进行分类，并输出JSON/CSV/Excel格式。

- Author: Song Chen
- Repository: lightme1020/Reportplatformautore_skills
- Version: 20260127235042
- Stars: 0
- Forks: 0
- Last Updated: 2026-02-06
- Source: https://github.com/lightme1020/Reportplatformautore_skills
- Web: https://mule.run/skillshub/@@lightme1020/Reportplatformautore_skills~brick-strength-recognition:20260127235042

---

---
name: brick-strength-recognition
display_name: 砖强度表格识别
description: 从砖强度检验表格（PDF/图像）中提取结构化字段，对表格类型进行分类，并输出JSON/CSV/Excel格式。
---
# Brick Strength Recognition Skill
## When to Use
Use this skill when users need to recognize brick strength inspection tables and extract structured fields from them.

## Execution Flow
1. Accept a PDF or image file
2. Classify the table type (if multiple templates exist)
3. Extract structured fields according to the target schema
4. Output JSON, CSV, or Excel

## Output Formats
- JSON: Structured JSON data
- CSV: Comma-separated values
- Excel: Excel workbook

## Field Schema Source
The field list and schema live in `references/fields.md`.
Read that file before extraction and follow the exact field names and normalization rules.
Examples of expected output formats are in `examples/`.

## How to Run
From the skill root:
- `python scripts/run.py batch_process.py <pdf_or_image_paths>`

## Template Notes
- Current skill covers one known template; more templates may be added later.
- In the current template, header keywords are fixed.
- If a future template is unclear, ask for clarification and provide a short guess with rationale.

## Notes
- Keep missing fields as `null` and record any ambiguities in a short `notes` field.