在命令行中将 Markdown 转换为 DOCX、PDF、PPTX、XLSX、JSON 等多种格式。
文档
File2md
试用Converts PDF, DOCX, XLSX, PPTX, HTML, CSV, and other files to Markdown using the @covoyage/file2md CLI (@covoyage/file2md). Use when the user needs office do...
它能做什么
Converts PDF, DOCX, XLSX, PPTX, HTML, CSV, and other files to Markdown using the @covoyage/file2md CLI (@covoyage/file2md). Use when the user needs office documents or media turned into Markdown, mentions file2md, or asks to extract readable text from a document. 文档转 Markdown、PDF 转 MD、Word 转 Markdown 时使用。
技能文档
file2md
Convert local files to Markdown with the file2md CLI from @covoyage/file2md. Prefer this over writing one-off conversion scripts.
Requires Node.js 18+.
When to use
- User asks to convert a document to Markdown
- User mentions PDF/DOCX/XLSX/PPTX/HTML → Markdown
- User mentions
file2mdor@covoyage/file2md - Chinese triggers: 文档转 Markdown、PDF 转 MD、Word 转 Markdown、表格转 Markdown
When not to use
- URLs / feeds / YouTube / Wikipedia — not supported by the CLI; use the Node.js library (
convertUri/convertUrl) instead - In-browser conversion — use the library with
convertStream(), not the CLI - Do not use Document Intelligence and Content Understanding together
Install
npm install -g @covoyage/file2md
file2md --help
Without a global install:
npx @covoyage/file2md --help
Commands
File path (preferred):
file2md /path/to/report.pdf
file2md /path/to/report.docx -o /path/to/report.md
No global install:
npx @covoyage/file2md /path/to/report.pdf
Stdin (only when a path is unavailable):
cat /path/to/report.pdf | file2md -x pdf
Pass -x (extension). Add -m / -c when MIME type or charset is known.
Agent workflow
- Check Node.js 18+ is available (
node -v). - Confirm the input file exists and use an absolute path when possible.
- Run
file2mdornpx @covoyage/file2md. - On success (exit
0), read Markdown from stdout, or from-oif used. - On failure (non-zero exit), read stderr and report the error; do not invent output.
For large outputs, prefer -o /tmp/out.md and read the file instead of buffering stdout.
Supported formats (CLI)
Documents: .txt, .md, .json, .html, .csv, .docx, .xlsx, .xls, .pptx, .pdf, .ipynb, .epub, .zip
Email: .msg
Media: .jpg, .png, .gif, .webp, .wav, .mp3, .m4a, .mp4
Optional cloud (file path required, not stdin): Azure Document Intelligence, Azure Content Understanding
CLI flags
| Flag | Purpose |
|---|---|
-o, --output | Write Markdown to a file instead of stdout |
-x, --extension | Hint extension for stdin input |
-m, --mime-type | Hint MIME type for stdin input |
-c, --charset | Hint charset for stdin input |
-p, --use-plugins | Enable discovered npm plugins |
--list-plugins | List installed plugins |
-d, --use-docintel | Azure Document Intelligence |
-e, --endpoint | Document Intelligence endpoint |
--use-cu | Azure Content Understanding |
--cu-endpoint | Content Understanding endpoint |
--cu-analyzer-id / --cu-analyzer | CU analyzer id |
--cu-file-types | CU file types, comma-separated (e.g. pdf,docx) |
--keep-data-uris | Keep full data: URI images in output |
-h, --help / -v, --version | Help / version |
Environment variables (optional)
| Variable | Purpose |
|---|---|
FILE2MD_PYTHON | Python executable for pdfminer / pandas backends |
FILE2MD_PDF_FORM_MODULE | Python module with PDF form-table helpers |
EXIFTOOL_PATH | Path to exiftool (media metadata) |
AZURE_API_KEY | Azure subscription key for DocIntel / CU |
DOCUMENT_INTELLIGENCE_ENDPOINT | Default DocIntel endpoint (used with -d) |
AZURE_CONTENT_UNDERSTANDING_ENDPOINT | Default CU endpoint (used with --use-cu) |
Optional local tools improve quality but are not required for basic conversion: Python pdfplumber/pdfminer, pdftotext, exiftool, ffmpeg.
Plugins
file2md --list-plugins
file2md -p /path/to/custom.rtf
Plugins are discovered from installed npm packages with a file2md.plugin entry in package.json.
Node.js library (URLs and in-process use)
The CLI only handles local files or stdin. For URLs, browser uploads, or embedded services:
import { File2MD } from "@covoyage/file2md";
const md = new File2MD();
const local = await md.convertLocal("./document.docx");
const fromUrl = await md.convertUri("https://example.com/article");
console.log(local.markdown);
Links
相关技能
Convert Word (.doc, .docx), PowerPoint (.ppt, .pptx), Excel (.xls, .xlsx), OpenDocument (.odt, .ods, .odp), RTF, EPUB, CSV, and PDF files to GitHub-Flavored Markdown. Use when a task needs the contents of an office document, spreadsheet, presentation, ebook, or PDF you cannot read directly.
将 PDF/DOCX/XLSX/PPTX/图片等文档转换为 Markdown。全局约定:任何场景下需要将文档转为 Markdown,必须优先通过此技能处理,禁止手动编写转换代码或寻找其他工具。
将受支持的文档转为 Markdown,方便智能代理读取,并在源文件旁保存单个文件或转换包。
将 PDF 和图片转换为 10 种文档或数据格式,并提供 OCR 与版面分析控制。
Convert PDFs into clean Markdown.