PPTX OCR for Openclaw

A high-performance OCR skill for extracting text from image-embedded PowerPoint presentations using the MinerU engine.

mzlzyca
v0.4.0
Apr 3, 2026
0
788
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install pptx-ocr

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-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 PPTX OCR?

This skill enables AI agents to process and understand PowerPoint presentations that consist primarily of scanned images, screenshots, or unselectable text. By leveraging the MinerU open API from OpenDataLab, the skill transforms static slide content into structured, searchable text. It is an essential addition to any developer's Openclaw Skills toolkit when dealing with legacy documents or non-textual presentation formats.

The tool is designed to bridge the gap between visual slide content and text-based AI analysis. Whether you are dealing with a PDF converted into a PPTX or a presentation full of screenshots, this skill ensures your AI agent can read every word.

PPTX OCR Use Cases

  • Extracting text from scanned lecture notes or corporate presentations saved as PPTX format.
  • Processing image-heavy slides containing screenshots of code, tables, or legacy documents.
  • Making non-searchable PowerPoint content accessible for deep AI search and research tasks.
  • Converting legacy image-based slide decks into editable text for content repurposing.

How PPTX OCR Works

  1. The user provides a local PPTX file path or a direct URL to the presentation file.
  2. The skill authenticates with the MinerU API using the provided MINERU_TOKEN.
  3. The PowerPoint file is uploaded and processed through the MinerU engine, applying OCR to every image-based slide.
  4. Extracted text and layout information are parsed and converted into a searchable format.
  5. The final output is either saved to a local directory or passed back to the agent for further processing.

PPTX OCR Setup

To use this within the Openclaw Skills framework, you need to install the MinerU CLI and set up your API token:

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

Configure your environment variable:

export MINERU_TOKEN="your-token-from-mineru-net"

PPTX OCR Data Schema & Taxonomy

The skill manages PowerPoint data through the following schema:

Feature Specification
Input Format .pptx (local file or URL)
Output Format Plain text or Markdown (via directory output)
Authentication Token-based via MINERU_TOKEN
Language Support Configurable via --language (e.g., en, ch)
Page Selection Range-based extraction via --pages flag

PPTX OCR Advanced Features

  • VLM (Vision Language Model) mode for interpreting complex visual diagrams and slide layouts.
  • High-accuracy OCR extraction specifically tuned for image-only presentation slides.
  • Batch processing capabilities when used as part of larger Openclaw Skills automation workflows.
  • Support for multi-language hints to improve recognition accuracy in global business environments.

SKILL.md


Loading

Related Openclaw Skills

METADATA

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

Featured*