PPTX Parse for Openclaw

A high-performance utility for converting PowerPoint files into structured, hierarchical Markdown using the MinerU document intelligence engine.

mzlzyca
v0.4.0
Apr 3, 2026
0
827
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install pptx-parse

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install pptx-parse using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is PPTX Parse?

PPTX Parse is a technical skill designed to transform visual PowerPoint presentations into clean, structured Markdown data. By utilizing the advanced MinerU engine from OpenDataLab, it allows developers and automated agents to extract text, preserve slide hierarchies, and process presentation content as machine-readable text. This skill is essential for any Openclaw Skills workflow that requires document intelligence and data extraction from legacy or modern presentation formats.

Whether you are processing local files or remote URLs, PPTX Parse provides the flexibility to handle high-volume document pipelines. It bridges the gap between static slides and actionable data, enabling Large Language Models to interpret and analyze presentation structures with high precision.

PPTX Parse Use Cases

  • Automating the conversion of slide decks into searchable knowledge base articles.
  • Extracting structured outlines from presentations for Large Language Model processing.
  • Creating text-based summaries of corporate presentations within Openclaw Skills automation flows.
  • Converting visual tables and formulas from slides into clean Markdown format for data analysis.

How PPTX Parse Works

  1. The user provides a path to a local .pptx file or a direct URL to the presentation.
  2. The skill utilizes the mineru-open-api to initiate either a quick flash-extract or a comprehensive extraction process.
  3. The document intelligence engine analyzes the internal structure of the slides to identify headings, body text, and layout hierarchy.
  4. Structured Markdown is generated, ensuring that the logical flow of the original presentation is maintained.
  5. The output is delivered via stdout or saved to a specified directory, ready for integration into the broader Openclaw Skills environment.

PPTX Parse Setup

To get started with PPTX Parse, install the necessary dependencies via npm or Go:

npm install -g mineru-open-api
# Alternatively, for macOS/Linux users:
go install github.com/opendatalab/MinerU-Ecosystem/cli/mineru-open-api@latest

For advanced features requiring full extraction, configure your authentication token:

export MINERU_TOKEN="your-api-token-here"

PPTX Parse Data Schema & Taxonomy

The skill produces a structured output that maps presentation elements to Markdown syntax. Below is how the data is organized:

Feature Description
Input Formats .pptx files (local or URL)
Output Format Hierarchical Markdown (.md)
Slide Delimiters Preserves page breaks and slide boundaries
Complex Elements Tables and formulas are converted to Markdown-compatible representations
Assets Optional directory output for extracted media and images

PPTX Parse Advanced Features

  • Support for flash-extract mode which allows for rapid parsing of files up to 10MB without an API token.
  • Granular page selection using the --pages flag to extract specific content ranges.
  • Language-specific optimization hints via the --language parameter to improve OCR and parsing accuracy.
  • Seamless integration with cloud-based document intelligence for complex formula and table recognition.
  • Flexible output management allowing for direct pipe-based processing or file-system storage.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Requires
Bins mineru-open-api
Github Stars: 0
forks: 0

Featured*