Word + Markdown · Context Board

Word to Markdown,
check it first.

Import a .docx into a doc board, then download the page as Markdown: a # title, your headings, bold, italic, links, lists and tables. Fix anything on the page first, or let your agent read it.

Free for 3 days · then $12 a month · every widget included
PagesImported from Wordnow
  • Q3 reportQ3-report.docx1 page
  • Onboarding guideOnboarding.docx1 page
  • Employee handbookHandbook.docx3 pages
  • Pricing FAQPricing-FAQ.docx1 page
  • Meeting notesNotes-Oct.docx1 page
Tables brought overFrom the Word filesnow
7
Claude Codedocs · main1m
Working
Reading the Q3 report as Markdown
Read the onboarding guide

Downloaded as q3-report.md: a # title, then your headings, lists and tables. Each picture is a link to your board.

Today
  • Import the Q3 report
  • Check the table in section 2
  • Download it as Markdown
An example Context Board named Docs. Every number on it is example data.
ON ONE BOARD

Why a Word file is hard to reuse as Markdown

Word keeps a document's headings, lists and tables as formatting. Markdown writes them as plain marks: #, -, | and asterisks. A command-line converter like pandoc writes the whole file in one go, and you see what it did only afterward, in another app. Pictures are the other snag, because a Markdown file holds a picture as a link and the image has to live somewhere. And your agent still needs the file put where it can read it.

How to convert a Word document to Markdown

  1. Bring in your .docx At board.contextbolt.com, choose New board, then Import a document, and drop the file. On a doc board, use the Import pages button above the page list, or drop the file on the list. From Google Docs, choose File, Download, Microsoft Word (.docx) first.
  2. Check the page Before it imports, the preview counts the tables and pictures and lists anything it left out. Then the document opens as a page: fix a heading or a table cell on it, or ask your agent to. The import summary has an Undo import button too.
  3. Download the Markdown Open the page's ⋯ menu in the page list, or right-click the page, and choose Download as Markdown. You get one .md file named after the page, starting with its # title. Or leave it on the board and let your agent read it.

Why convert on Context Board

Structure that carries overHeadings, bold, italic, links, bulleted and numbered lists and tables come across. A Title, or a first Heading 1, becomes the page's title. Fonts, colors and underlining stay behind.
Check it before you downloadThe page is the source of the Markdown, so fix a heading or a table cell there, then download. The file opens with the page's # title, and each table is a Markdown table with its first row as the header.
Your agent can read itClaude, ChatGPT, Claude Code, Codex or any app connected over MCP reads the page as Markdown with read_page, finds it by its words with search_boards, and edits it with write_page.

Frequently asked questions

Something else on your mind? Get in touch.

How do I convert Word to Markdown for free?

Use pandoc, a free command-line converter. Its manual lists docx as an input format and gfm, GitHub-Flavored Markdown, as an output format, so one command does it: pandoc -f docx -t gfm report.docx -o report.md. For the pictures, add --extract-media=media, which the manual describes as "Extract images and other media contained in or linked from the source document to the path DIR". You get a .md file and a folder, and nothing to read or edit the result in. If you want to check it, fix it and let your agent read it, import the .docx into a doc board on Context Board and download the Markdown from there.

What does the Markdown file contain?

The page's title as a # heading, then the page in order. Headings are #, ## and ###. Bold and italic are marked with asterisks, links keep their address, bullets are - lines and numbered lists are 1. lines. A table is a Markdown table whose first row is the header. Characters Markdown would read as formatting, like asterisks and square brackets, get a backslash in front.

Are the pictures in the Markdown file?

No. The file doesn't carry the image files. Each picture is an image line pointing to its address on your board, with its alt text from Word as the caption when it had one. The board keeps PNG, JPEG, GIF and WebP pictures, up to 60 in one import. A picture it can't keep is left out, and the preview counts it. If you need the image files themselves, take them from the original .docx.

What does the Word import leave out?

Fonts, text sizes and colors, and underlining. The reader turns a document into its structure rather than copying its look. Comments in the margin are not brought over. Heading 4 and below come in as Heading 3, because a page has three heading levels. An old .doc file is left out with a note to save it as .docx first. A picture of a kind the board can't keep, or one only linked from the file instead of inside it, is left out too.

Does it download the whole document, or one page?

One page. A Word file becomes a page, and Download as Markdown writes that page. A document too long for one page goes in parts, called Part 2, Part 3 and so on, each a page inside the first and each linking to the next, so each part downloads as its own file. Pages inside a page are not in its file. Several Word files make several pages. Your agent can read them all.

Can my agent read the Word file as Markdown?

Yes. Once the Word file is a page, an agent connected to your board over MCP, like Claude, ChatGPT, Claude Code, Codex, Cursor or VS Code, reads it as Markdown with read_page. search_boards finds pages by their words, and write_page adds to a page or replaces part of it. See how your agent works with the board.

What does it cost?

Context Board is free for 3 days, then $12 a month, with every board type and widget included. Doc is one of them, next to Sheet, Slides and Plan.

Your business, on one board

Free for 3 days, then $12 a month. One plan with every connection, widget and theme.

Start free trial See everything Context Board does