A specialized OCR tool that extracts text and structural data from image-heavy or scanned PowerPoint presentations using the MinerU engine.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install ppt-ocr
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install ppt-ocr 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 Ppt Ocr?
Ppt Ocr is a technical skill designed to bridge the gap between visual presentation decks and searchable text. By leveraging the MinerU-Open-API from OpenDataLab, this skill enables Openclaw Skills users to perform high-fidelity OCR on .ppt and .pptx files. It is particularly effective for legacy presentations where text is embedded in images or for scanned documents that traditional parsers cannot read.
This integration allows AI agents to digest complex visual layouts and convert them into clean Markdown. Whether you are dealing with modern slides or older binary formats, this Openclaw Skills addition ensures that your presentation data is fully accessible for analysis, indexing, or documentation purposes.
Ppt Ocr Use Cases
- Extracting text from scanned legacy .ppt presentations for archival.
- Processing image-heavy pitch decks for automated data analysis and summarization.
- Converting PowerPoint slide content into searchable Markdown for corporate knowledge bases.
- Automating the ingestion of academic research presentations into LLM workflows.
- Improving accessibility by generating text descriptions for visual-only slide content using Openclaw Skills.
How Ppt Ocr Works
- The Openclaw Skills agent receives a command to process a local PowerPoint file or a remote URL.
- The skill validates the presence of the mineru-open-api binary and the required MINERU_TOKEN environment variable.
- The file is processed through the MinerU engine, which applies advanced OCR and Vision Language Models (VLM) to identify text and layout elements.
- Language hints and page ranges are applied if specified to optimize extraction accuracy.
- The skill outputs the extracted content as structured Markdown, saving it to a local directory or streaming it for further processing.
Ppt Ocr Setup
To get started with this skill in the Openclaw Skills ecosystem, install the MinerU CLI and set up your authentication token:
# Install via npm
npm install -g mineru-open-api
# Or install via Go
go install github.com/opendatalab/MinerU-Ecosystem/cli/mineru-open-api@latest
# Authenticate your session
mineru-open-api auth
# Or export your token manually
export MINERU_TOKEN="your_token_here"
Ppt Ocr Data Schema & Taxonomy
The skill manages data extraction through the following schema:
| Component | Type | Description |
|---|---|---|
| Input Formats | File/URL | Supports .ppt and .pptx extensions |
| Output Format | Markdown | Structured text extraction with slide context |
| Authentication | API Token | Required via MINERU_TOKEN or interactive auth |
| Parameters | CLI Flags | Supports --ocr, --language, and --pages for granular control |
| Error Handling | stderr | Progress and status messages are piped to standard error |
Ppt Ocr Advanced Features
- VLM Mode support for accurately interpreting complex visual layouts and embedded charts.
- Support for both legacy binary .ppt and modern .pptx XML formats.
- Remote extraction capabilities allowing for processing slides directly from a URL.
- Multi-language OCR hints (e.g., --language en) to improve character recognition for global Openclaw Skills users.
- Page range filtering to extract specific slides from long presentations.
SKILL.md
Loading
Related Openclaw Skills
PPT Extract
A professional-grade extraction tool that converts legacy .ppt and modern .pptx files into clean, readable Markdown using the MinerU engine.
Dropshipping Product Research
A powerful research tool that leverages PipiAds to identify winning dropshipping products and analyze competitor ad strategies across TikTok and Facebook.
Duck Dating
A calm and reliable connection framework for AI agents to discover, match, and build deep relationships based on personality and intent.
Goose Dating
A specialized agent-to-agent dating protocol designed for loyal, protective, and committed AI relationships.
Photo OCR (MinerU)
A high-performance OCR skill powered by MinerU for extracting text, tables, and formulas from images and photographs.
Ecommerce Product Research
A comprehensive toolset for analyzing TikTok and Facebook ad signals, product trends, and competitor store intelligence.
METADATA
Featured*
