Deprecated
MarkItDown
Convert documents, images, audio, and web content to LLM-optimized Markdown
Document Processing
About
DEPRECATED: this server is no longer supported by the Obot MCP Catalog. No direct replacement is currently available. Convert documents, images, audio, and web content to clean, structured Markdown optimized for LLM consumption.
Features
- Multi-Format Support: Convert PDF, PowerPoint, Word, Excel, Images (with OCR), Audio (with transcription), HTML, CSV, JSON, XML, ZIP files, EPubs, and YouTube URLs
- Structure Preservation: Maintains important document structure including headings, lists, tables, and links in Markdown format
- LLM-Optimized Output: Markdown output is designed specifically for large language model processing and text analysis pipelines
- Metadata Extraction: Extracts EXIF metadata from images and audio files
- Web Content: Fetch and convert web pages and online documents via HTTP/HTTPS URLs
- Token Efficient: Markdown conventions are highly token-efficient for LLM applications
What you’ll need to connect
No Setup Required: MarkItDown works out-of-the-box with no API keys or configuration needed. Simply provide a URI (http://, https://, file://, or data:) to convert content to Markdown.
Configuration
Container runtime
- Image
ghcr.io/obot-platform/mcp-images/markitdown-mcp:0.0.1a4-obot2- Port
8099- Path
/- Args
markitdown-mcp