Miaoda Doc Parse

Use when user needs to parse, extract, or read content from documents (PDF, Word, Excel, PPT, etc.) Converts documents to Markdown via IDP (Intelligent Docum...

Security Scan

The skill's instructions require running a 'miaoda-studio-cli' tool and accessing local files/URLs, but the package declares no required binaries or install instructions and has no source/homepage — the pieces are incomplete and mismatched.

Purpose & Capability

The SKILL.md describes converting documents to Markdown via an IDP and gives examples that invoke the CLI command 'miaoda-studio-cli doc-parse'. However the skill metadata lists no required binaries, no install spec, and there is no homepage or source URL. Requiring a CLI to perform parsing is reasonable, but the skill fails to declare or provide that dependency or guidance on how to obtain it.

Instruction Scope

Instructions are narrowly scoped to parsing local files or downloadable URLs (expected for a doc-parse tool) and reference related skills (web-crawl, feishu). They do instruct reading local file paths and fetching remote URLs, which is appropriate for the purpose but implies the agent will access local files and network resources. The SKILL.md does not instruct the agent to read unrelated system files or secrets.

Install Mechanism

There is no install specification despite explicit CLI usage in the instructions. Because SKILL.md assumes 'miaoda-studio-cli' is present, the skill is incomplete: it doesn't declare a required binary or provide an install source (package, URL, or repo). The absence of a trustworthy install origin (homepage/source) increases risk if a user were to manually install an unknown CLI.

Credentials

The skill declares no environment variables or credentials and the documentation does not ask for secrets. This is proportionate to a document-parsing helper.

Persistence & Privilege

The skill is not always-enabled, is user-invocable, and does not request persistent platform privileges. Nothing in the metadata indicates it would modify other skills or agent-wide settings.

What to consider before installing

This skill appears to be an instruction wrapper for a CLI named 'miaoda-studio-cli', but it does not declare that binary or provide install/source information. Before installing or using it: (1) verify where 'miaoda-studio-cli' comes from (official repo or vendor) and only install from a trusted source; (2) confirm the CLI's permissions and what network endpoints it contacts; (3) be aware that using the skill will allow the agent to read local files and fetch remote documents — avoid feeding sensitive documents until you trust the CLI; (4) ask the author or publisher for an install spec or homepage so the missing dependency and provenance are resolved.

Document Parse

通过 miaoda-studio-cli doc-parse 将文档解析为 Markdown 格式（基于 IDP 智能文档解析）。

Quick Reference

参数	说明	必需	默认值
--file	文档路径或 URL	是	-
--output, -o	输出格式: text/json	否	text

支持的文档格式

格式	扩展名	典型场景
PDF	.pdf	报告、论文、合同
Word	.doc, .docx	文档、方案
PowerPoint	.pptx	演示文稿
Excel	.xlsx	表格数据
CSV	.csv	结构化数据
纯文本	.txt, .md	文本文件
网页	.html	HTML 页面

使用示例

# 解析本地 PDF 文件
miaoda-studio-cli doc-parse --file report.pdf

# 解析远程文档并输出 JSON
miaoda-studio-cli doc-parse --file https://example.com/document.docx --output json

# 解析 Excel 表格
miaoda-studio-cli doc-parse --file data.xlsx --output json

使用场景决策

需要获取文档内容
├─ 文档是本地文件或可下载 URL → miaoda-studio-cli doc-parse
├─ 文档是网页（需要抓取）→ miaoda-studio-cli web-crawl（见 miaoda-web-fetch skill）
└─ 文档是飞书云文档 → 使用 feishu skill

Common Mistakes

错误	正确做法
用 miaoda-studio-cli doc-parse 解析网页 URL	网页内容用 miaoda-studio-cli web-crawl（见 miaoda-web-fetch skill）
不检查文件格式是否支持	确认文件扩展名在支持列表中
需要结构化数据但未加 --output json	需要程序处理时加 --output json
文件路径包含空格未加引号	路径有空格时用引号包裹：--file "my file.pdf"
解析飞书云文档用 doc-parse	飞书云文档使用 feishu skill 通过 API 读取