“Fill our work instruction template from this Loom recording of the pivot table setup.”
- app_run · video-to-word-template
- app_run_status
- app_run_result
The filled Word file to download, with a list of the fields the video couldn't answer.
Docsie MCP server · Word templates
Give your AI assistant a recording and the Word form your team already uses for work instructions, SOPs or procedures. Docsie fills every field the video supports, writes out each step with screenshots, and lists the fields left for you.
Runs the same Video to Word Template app as Docsie. Uses AI credits.
https://app.docsie.io/mcpWorks in Claude, ChatGPT, Claude Code, Cursor, Codex and any MCP client.
A plain Word work instruction, and the same template filled from a LibreOffice Calc recording with its steps and screenshots. The same app runs over MCP.
What you can ask
Your assistant runs Video to Word Template with the .docx you name or the template saved in your app settings. If the content is already in Docsie, fill_docx_template fills your template from it.
“Fill our work instruction template from this Loom recording of the pivot table setup.”
The filled Word file to download, with a list of the fields the video couldn't answer.
“Use QA-Form-v3.docx from my Docsie files as the template for this screen recording.”
The run fills that .docx instead of the template saved in your app settings.
“Fill the SOP template with the steps from the guide Docsie wrote from my video earlier.”
A Word file in your template's layout, filled from that guide.
Watch it work
Walkthroughs recorded in the Docsie app. Over MCP, your assistant runs the same app and gets the same results back in the chat.
Real results
Results from the Video to Word Template app on a four-minute LibreOffice Calc recording and a plain Word form. Over MCP, the same app runs and your assistant reads back the result.
Labels such as Purpose, Scope, Required equipment, Step 1 and Step 2, and a revision history table. No placeholders were added to the template.

The template had two step lines and the video showed twelve steps, so twelve were written, ten with a frame from the recording. The header and styles stay as they were.

The result names the template, the fields it filled and the fields left for you, with PDF, DOCX and Markdown downloads.

How it works
Save your .docx in the app settings once, or name a different Word file in your Docsie workspace for one run. A plain Word form with labels works.
Give a YouTube, Loom, Vimeo or Google Drive link, a direct video URL, or a video already in Docsie.
The result says which fields were filled and which are left for you, such as an owner or approver that a recording can't know.
Permissions and limits
Running Video to Word Template needs Workspace create and edit, which is off by default. fill_docx_template is part of the default Video to docs pack.
Every fill uses your organization's AI credits, the same as running the app in Docsie. The MCP server itself is on every plan.
Fields a video can't answer, such as a document number, an owner or safety notes, are left blank and listed for you to complete.
Video to Word Template →FAQ
On their own they can write text for you to paste into Word. Connected to the Docsie MCP server, they can run Docsie's Video to Word Template app or fill_docx_template, which fill your .docx and keep its layout and styles.
No. A plain Word form with labels such as Purpose, Scope and Step 1 works. Templates that already use placeholders are filled too.
The .docx you name for that run, or the Word template saved in your app settings. One of the two is needed.
YouTube, Loom, Vimeo and Google Drive links, direct video file URLs, and videos already in your Docsie workspace.
The MCP server is on every Docsie plan. Each fill uses AI credits, the same as in the app.
Docsie MCP server
Connect Claude, ChatGPT or Cursor to Docsie and fill your Word templates from recordings without leaving the chat.