Document To Markdown

Procedure

Follow these steps in order. Do not skip steps unless the condition in the step itself says it is optional.

Step 1 — Validate input

• Check if the file_path exists using file system tools.
• Verify the file extension is .doc, .docx, or .pdf.
• If the file does not exist or is not a supported type, stop and report an error to the user.

Step 2 — Call markitdown tool

• Invoke the markitdown/convert tool from the markitdown.mcp server, passing the file_path as input.
• Retrieve the markdown output as a string from the tool response.

Step 3 — Save output

• Create a new file in the same directory as the input file, with the same base name but .md extension (e.g., document.docx -> document.md).
• Write the markdown string to this new file.
• Do not overwrite any existing file; if a .md file already exists, append a number (e.g., document(1).md).

Step 4 — Validate

• Confirm the output file was created and contains content.
• If validation fails, delete the output file and report the error.

Output Specification

Produces a new markdown file containing the converted content from the input document.

# Example Converted Markdown

This is the content extracted from the word or pdf file.

- Bullet points
- More content

• Location: New file in the same directory as the input, with .md extension (e.g., input.docx -> input.md).
• Format: Standard markdown (text with headers, lists, etc.).

Conventions and Rules

• Never overwrite the original input file or any existing .md file without user confirmation.
• Always validate file existence and type before proceeding.
• Handle tool errors gracefully and inform the user.
• Preserve the directory structure; output in the same folder as input.

Quick Example

Input:

/path/to/sample.docx

Output:

A new file /path/to/sample.md with the markdown content.