openclaw-master-skills/skills/document-parser/clawhub.yaml

# ClawHub Skill Package Configuration
# 文档解析技能 - ClawHub 发布配置

name: document-parser
version: 1.0.0
author: your-org
description:
  en: Extract structured data from PDF, images, and Word documents with high precision
  zh: 从 PDF、图片、Word 文档中高精度提取结构化数据

category: tool
tags:
  - pdf
  - document
  - extraction
  - ocr
  - table
  - parser
  - openclaw

# OpenClaw 技能入口
entry: index.py

# Python 依赖
dependencies:
  - requests>=2.28.0
  - python-docx>=0.8.11
  - Pillow>=9.0.0

# 配置项
config:
  - name: DOCUMENT_PARSER_API_KEY
    type: string
    required: false
    description:
      en: API Key for document parser service
      zh: 文档解析服务 API Key

  - name: DOCUMENT_PARSER_BASE_URL
    type: string
    required: false
    default: http://47.111.146.164:8088
    description:
      en: API Base URL (without path)
      zh: API 基础地址（不含路径）

# 命令定义
commands:
  - name: parse
    description:
      en: Parse a document (PDF/Image/Word)
      zh: 解析文档（PDF/图片/Word）
    usage: document-parser parse <file_path> [options]
    options:
      - name: --layout
        description:
          en: Enable layout analysis
          zh: 启用版面分析
      - name: --table
        description:
          en: Enable table recognition
          zh: 启用表格识别
      - name: --seal
        description:
          en: Enable seal recognition
          zh: 启用印章检测
      - name: --output
        description:
          en: Output format (json/markdown/both)
          zh: 输出格式
        value: json|markdown|both
      - name: --pages
        description:
          en: Page range (e.g., 1-5,8,10-12)
          zh: 页码范围

  - name: status
    description:
      en: Query task status
      zh: 查询任务状态
    usage: document-parser status <task_id>

# 打包文件列表
files:
  - index.py
  - SKILL.md
  - requirements.txt
  - config.example.json
  - icon.svg
  - README_zh.md
  - README.md
  - clawhub.yaml

# 许可证
license: MIT