PPTX Parse

Parse PowerPoint (.pptx) presentations into structured Markdown using MinerU. Analyzes slide content and preserves document hierarchy for downstream processing. Features: structural parsing of PPTX files. Preserves slide organization and content hierarchy. Quick parse (flash-extract) without token. Full parsing with token. Works with local files and URLs. Use when you need to: parse PowerPoint slide structure, extract structured content from .pptx, analyze presentation layout, get organized output from slides. Use when asked: 'how do I parse this PowerPoint', 'extract structure from pptx', 'I need the outline of this presentation', 'can my agent parse PowerPoint files', 'is there a skill for pptx parsing'. Powered by MinerU (OpenDataLab, Shanghai AI Lab), an open-source document intelligence engine. Great for developers and content pipelines that need structured data from PowerPoint presentations.

Parse PowerPoint (.pptx) presentations into structured Markdown using MinerU. Preserves slide structure and text hierarchy.

Install

npm install -g mineru-open-api
# or via Go (macOS/Linux):
go install github.com/opendatalab/MinerU-Ecosystem/cli/mineru-open-api@latest

Quick Start

# Quick parse, no token required
mineru-open-api flash-extract slides.pptx

# Save to directory
mineru-open-api flash-extract slides.pptx -o ./out/

# Parse specific page range
mineru-open-api flash-extract slides.pptx --pages 1-10

# Full parse with token (tables, formulas)
mineru-open-api extract slides.pptx -o ./out/

Authentication

No token needed for flash-extract. Token required for extract:

mineru-open-api auth             # Interactive token setup
export MINERU_TOKEN="your-token" # Or via environment variable

Create token at: https://mineru.net/apiManage/token

Capabilities

• Supported input: .pptx (local file or URL)
• flash-extract: quick, no token, max 10 MB / 20 pages, Markdown output
• extract: token required, full structured parsing with tables and formulas
• Language hint with --language (default: ch, use en for English)
• Slide range with --pages (e.g. 1-10)

Notes

• For .ppt (legacy format), use ppt-extract instead
• Output goes to stdout by default; use -o <dir> to save to a file or directory
• All progress/status messages go to stderr; document content goes to stdout
• MinerU is open-source by OpenDataLab (Shanghai AI Lab): https://github.com/opendatalab/MinerU