How to Any File to Markdown: Automating Knowledge Flow into Growi

Tips & Tricks
, , , , ,
1

Hi everyone!:open_hands:

Document management and automation remain overarching challenges in AI adoption.
In Japan in particular, the strong cultural reliance on Excel for document management has become a bottleneck for effective AI utilization.

:bullseye: Objectives

  • Balancing document conversion for AI with human-readable documentation

  • Standardizing into Markdown format for easier AI processing

  • Building a knowledge base optimized for human readability

:white_check_mark: Results

  • Automated workflow confirmed: Store files in S3 → Convert to Markdown → Save to Growi

  • Supported formats: DOCX, PDF, TXT, CSV, XLSX

:counterclockwise_arrows_button: Processing Flow

  • Batch processing: Automatically detect and convert all files in the input/ folder in S3

  • AI-powered conversion: High-accuracy Markdown transformation using Docling MCP and Gemini AI

  • Auto-formatting: Optimize heading hierarchy, tables, and code blocks

  • Growi integration: Automatically save converted results into the knowledge base using Growi MCP and Gemini AI

:light_bulb: Achievements

  • For AI: Structured conversion into Markdown

  • For humans: Improved information access through Growi’s search and browsing functions

  • For efficiency: Zero manual work for document standardization

:warning: Current Limitations

  • Files containing images are not yet supported (under consideration)

image
image1504×337 100 KB