MDC - Markdown Document Converter

AI-powered tool that converts presentations, documents, and files to Markdown format. Extract content from PowerPoint, PDF, Word, and other document types using advanced AI text extraction and conversion technology.

What is MDC?

MDC (Markdown Document Converter) is a powerful AI-driven tool that extracts content from presentations (PowerPoint, PPTX, PPT, PPSX), documents (PDF, DOC, DOCX), and other file formats and converts them into clean, organized Markdown files. It leverages advanced AI models to turn images into text, automatically organize headings and images, generate comprehensive table of contents, and maintain document structure during conversion.

Preview

MDC extracts content from presentations (PowerPoint, PPTX, PPT, PPSX), documents (PDF, DOC, DOCX), and other file formats and converts them into clean, organized Markdown files. It leverages advanced AI models to turn images into text, automatically organize headings and images, generate comprehensive table of contents, and maintain document structure during conversion.

Key Features

Multi-format Support: Convert PowerPoint presentations, PDF documents, Word files, and more
AI-Powered Extraction: Use OpenAI, Gemini, Ollama, Together, or OpenRouter APIs for intelligent text extraction
Image Processing: Automatically extract and convert images to text using AI vision models
Smart Organization: Automatically detect and organize headings, create table of contents (work in progress)
Attachment Support: Extract and process embedded files and media
Customizable: Configure API settings, processing parameters, and output formats
Cross-Platform: Works on Windows, macOS, and Linux
Obsidian Integration: Available as a plugin for Obsidian knowledge management

Quick Install Guide

Prerequisites

System Requirements:

Node.js (18+)
Poppler-utils (for PDF processing)
LibreOffice (for document conversions)
Internet connection for AI API access

Install commands:

Windows:

# Install Node.js
winget install OpenJS.NodeJS.LTS

# Install Poppler
choco install poppler

# Install LibreOffice
winget install TheDocumentFoundation.LibreOffice

macOS:

# Install Node.js
brew install node

# Install Poppler
brew install poppler

# Install LibreOffice
brew install --cask libreoffice

Linux (Ubuntu/Debian):

# Install Node.js
curl -fsSL https://deb.nodesource.com/setup_18.x | sudo -E bash -
sudo apt install -y nodejs

# Install Poppler and LibreOffice
sudo apt install poppler-utils libreoffice

Installation Methods

Method A: Global CLI Installation

# Clone the repository
git clone https://github.com/dralkh/mdc.git
cd mdc

# Install dependencies and build
npm install && npm run build-all

# Install system-wide (makes mdc command available globally)
npm link

# Verify installation
mdc --version

Method B: Obsidian Plugin Installation

# Install and test obsidian plugin
npm run install-plugin

# Install plugin to specific directory (non-interactive)
npm run install-plugin -- --d "/path/to/obsidian/plugins"
# or
npm run install-plugin -- --directory "/path/to/obsidian/plugins"

Method C: Local Development Setup

# For development without global installation
npm install && npm run build-all
# Use via npm script: npm run mdc -- "file.pptx" --md

Configure AI API Access

MDC supports multiple AI providers. Choose one or more for your conversion needs.

Setup API Keys

Create a .env file in the project folder by copying the example:

cp .env.example .env

Then, edit the .env file to add your API keys:

# OpenRouter (Recommended - supports multiple models)
OPENROUTER_API_KEY=your_openrouter_key_here

# OpenAI (GPT-4, GPT-3.5)
OPENAI_API_KEY=your_openai_key_here

# Google Gemini
GEMINI_API_KEY=your_gemini_key_here

# Together AI
TOGETHER_API_KEY=your_together_key_here

# Ollama (Local models)
OLLAMA_API_KEY=your_ollama_key_here

# OpenAI-Compatible API (Optional)
# Use this to connect to a custom or local OpenAI-compatible API endpoint
MDC_OPENAI_BASE_URL=http://localhost:8080/v1

Supported AI Models

Ensure the model supports vision and text

Basic Usage

MDC provides flexible command-line interface for converting various document formats to Markdown.

Quick Start Examples

# Convert PowerPoint to Markdown (most common use)
npm run mdc -- "presentation.pptx" --md --api openrouter

# Convert PDF to Markdown
npm run mdc -- "document.pdf" --md --api openrouter

# Convert Word document to Markdown
npm run mdc -- "document.docx" --md --api openrouter

# Include images/attachments
npm run mdc -- "presentation.pptx" --md --ma --api openrouter

# Generate table of contents
npm run mdc -- "document.pdf" --md --table --api openrouter

# Convert with custom API key
npm run mdc -- "file.pptx" --md --api openai --api-key "your-key"

Advanced Usage Patterns

# Batch conversion of multiple files
for file in *.pptx; do
  npm run mdc -- "$file" --md --api openrouter
done

# Convert with verbose output for debugging
npm run mdc -- "complex.pdf" --md --verbose --api openrouter

# Use custom configuration
npm run mdc -- "document.docx" --md --config "./custom-config.yaml" --api gemini

Command Line Options

Input Arguments

input-file - Path to the input file (supports: PPTX, PPT, PPSX, PDF, DOC, DOCX)

Output Options

-V, --version - Output the version number and exit
--md - Convert the extracted content into a Markdown (.md) file
--mr - Minimal output mode (only extract basefilename_markdown_text.md)
--table - Include a table of contents in the generated markdown
--headings - Auto-update heading hierarchy based on detected TOC structure (work in progress)

AI Configuration

--api <provider> - API provider to use (openrouter, openai, ollama, together, or gemini) (default: "openrouter")
--api-key, -k <key> - API key for the selected provider
--token <limit> - Maximum number of tokens per chunk for processing
--requests-per-minute <number> - Maximum number of API calls per minute

Processing Options

--ma - Include attachments extraction and processing
--identical-image-threshold <number> - Threshold for discarding identical images during PDF media extraction
--toc-file <path> - Optional: Path to the TOC markdown file
--config <path> - Optional: Path to custom config.yaml file

Debugging Options

--verbose - Enable verbose output, creating intermediate files
-h, --help - Display help for command

Usage Examples

# Basic conversion with default settings
mdc "presentation.pptx" --md

# Use specific API provider
mdc "document.pdf" --md --api openai

# Include attachments and table of contents
mdc "presentation.pptx" --md --ma --table

# Use custom API key inline
mdc "file.pdf" --md --api openrouter --api-key "your-key-here"

# Verbose mode with custom config
mdc "document.docx" --md --verbose --config "./custom-config.yaml"

Performance Optimization

For Large Documents

# Process with smaller token chunks
mdc "large_document.pdf" --md --token 2000 --api openrouter

# Enable verbose mode to monitor progress
mdc "complex_presentation.pptx" --md --verbose --api openai

# Use local models for faster processing (offline)
mdc "document.docx" --md --api ollama

Management

Monitor token usage with --verbose flag
Choose appropriate models for your needs
Consider batch processing for multiple files
Use local models (Ollama) when possible to reduce costs

Use Cases

Content Migration

Convert legacy PowerPoint presentations to Markdown
Migrate Word documents to modern documentation systems
Transform PDF reports into editable Markdown format

Knowledge Management

Build personal knowledge bases from existing documents
Create structured notes from presentation slides
Integrate with Obsidian, Logseq, or other PKM tools

Academic and Research

Extract text from research papers and publications
Convert lecture slides to study notes
Archive and organize educational materials

Business and Productivity

Convert meeting presentations to action items
Transform technical documentation to Markdown
Create knowledge bases from company materials

Integrations

Obsidian Plugin

MDC is available as a third-party plugin for Obsidian:

Seamless integration with Obsidian vault
Right-click context menu for quick conversions
Customizable settings within Obsidian
Automatic file organization and linking

Support & Contributing

If you encounter any issues, have feature requests, or would like to contribute, please let me know.

License

MIT License - see LICENSE file for details.

Name		Name	Last commit message	Last commit date
Latest commit History 9 Commits
src		src
.env.example		.env.example
LICENSE		LICENSE
README.md		README.md
build-cli.js		build-cli.js
config.yaml		config.yaml
esbuild.cli.config.mjs		esbuild.cli.config.mjs
esbuild.config.mjs		esbuild.config.mjs
install.ts		install.ts
main.js		main.js
main.ts		main.ts
manifest.json		manifest.json
package-lock.json		package-lock.json
package.json		package.json
styles.css		styles.css
tsconfig.json		tsconfig.json
update_settings.js		update_settings.js
version-bump.mjs		version-bump.mjs
versions.json		versions.json

License

dralkh/mdc

Folders and files

Latest commit

History

Repository files navigation

MDC - Markdown Document Converter

What is MDC?

Preview

Key Features

Quick Install Guide

Prerequisites

Installation Methods

Method A: Global CLI Installation

Method B: Obsidian Plugin Installation

Method C: Local Development Setup

Configure AI API Access

Setup API Keys

Supported AI Models

Basic Usage

Quick Start Examples

Advanced Usage Patterns

Command Line Options

Input Arguments

Output Options

AI Configuration

Processing Options

Debugging Options

Usage Examples

Performance Optimization

For Large Documents

Management

Use Cases

Content Migration

Knowledge Management

Academic and Research

Business and Productivity

Integrations

Obsidian Plugin

Support & Contributing

License

About

Topics

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases 1

Packages 0

Languages

Packages