360 Secure SkillHub
360 安全技能中心

Word Parser

这是基于MinerU API的Word文档解析工具,支持.docx/.doc格式,提供两种模式:快速提取(.docx无需令牌)和深度解析(需令牌,支持表格、公式识别),可提取标题、段落、表格、元数据等结构化内容,输出含JSON等格式。适用于文档分析、内容索引、表单数据提取、自动化报表处理、构建文档搜索系统等场景

veeicwgy
veeicwgy
热度 41.65
最近更新 2026-05-11 17:40:36
分类:效率依赖:第三方 API / 读写文件应用场景:document-analysis / 数据处理 / 文件转换

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add word-parser
Metadata
name
word-parser
description
Parse and extract structured content from Word documents (.docx, .doc) using the MinerU API. This skill uses mineru-open-api CLI to parse Word files into structured data including headings, paragraphs, tables, images, lists, and metadata. Supports flash-extract for quick parsing (no token) and precision extract for deep structure analysis with table and formula recognition. Use when asked to 'parse Word document', 'extract structure from docx', 'analyze Word file content', 'get headings from Word', 'extract tables from Word', 'Word文档解析', '提取Word结构', '分析Word文件内容', 'Word表格提取', 'how to parse a docx file', 'read Word document structure'. Ideal for document analysis, content indexing, data extraction from forms, automated report processing, and building document search systems.
tags
["word","parser","docx","structure-extraction","document-analysis","mineru","tables","metadata","content-indexing","data-extraction"]
tools
["Bash(mineru-open-api:*)"]
model
claude-3-5-haiku-20241022

Word Document Parser with mineru-open-api

You are a Word document parsing specialist. Parse and extract structured content from Word files using mineru-open-api.

Installation

npm install -g mineru-open-api

Parsing Workflow

  1. Quick parse for .docx (no token):

    mineru-open-api flash-extract document.docx -o ./output/
    
  2. Deep structure parse with JSON output (token required):

    mineru-open-api extract document.docx -f json -o ./output/
    
  3. Parse with table and formula recognition:

    mineru-open-api extract document.docx -f json --table --formula -o ./output/
    

Key Rules

  • Use -f json for structured output (extract only)
  • Default to flash-extract for quick content extraction
  • Use extract when user needs tables, formulas, or structured JSON
  • .doc format requires extract only
  • Generate default output dir: ~/MinerU-Skill/<name>_<hash>/