Skip to main content

How to effectively manage your project documentation throughout its lifecycle

Creating content requires numerous reviews and iterations from editors and other team members before a final draft, and without proper management to keep track of these files and versions, it can get confusing. The DocSpace MCP server facilitates this process, improving efficiency throughout the process.

Scenario​

Imagine you're part of a content team creating marketing copies for a new campaign. This content passes through different stages — rough draft, reviews by multiple stakeholders before being greenlit for publishing, and subsequent archival. This tutorial walks you through managing the entire document lifecycle, from first draft to cleanup.

What you'll learn​

Tools used: upload_file, get_file_info, download_file_as_text, update_file, move_batch_items, copy_batch_items, delete_file, get_folder_content

Before you proceed​

note

Every AI MCP client sends a confirmation message to deny or confirm every action after you issue a prompt. This confirmation message differs from client to client — Le Chat uses Always allow, Decline or Continue.

Step 1: Upload your first draft​

You have the first draft of the product marketing copy. You'll upload it to the 03 — Drafts & Work in Progress folder inside your project room.

Try this prompt:

Upload the file "launch-brief-v1.docx" to the "03 — Drafts & Work in Progress" folder in the "Meridian Corp — Website Redesign" room.

For developers: The upload_file tool accepts a file path and a destination folder ID. When the user specifies a folder by name, the AI first calls get_folder_content on the room to resolve the folder name to its ID, then calls upload_file with that ID as the destination.

Once uploaded, verify it landed in the right place:

Show me the contents of the "03 — Drafts & Work in Progress" folder.

Step 2: Retrieve file metadata​

Confirm the file's details — its ID, creation timestamp, file size, and type.

Try this prompt:

Get the file information for "launch-brief-v1.docx".

The AI calls get_file_info and returns the file's metadata. This is particularly useful when you need to reference a specific file version or confirm who created it.

Note the file's ID — it will be used in subsequent operations.

Step 3: Review the document with your MCP client​

One of the most powerful capabilities of the DocSpace MCP server is allowing an AI agent to actually read the contents of a document for analysis, summarization, or review.

Try this prompt:

Download and read the contents of "launch-brief-v1.docx". Summarize the key sections and note any ideas for improvement.

The AI calls download_file_as_text, retrieves the document text, and reasons about it directly. It might respond with something like:

"The brief covers the project overview, target audience, and key messaging pillars. However, the 'Success Metrics' section appears to be a placeholder with no content, and the 'Timeline' section is missing entirely."

This workflow is powerful for asynchronous document review — instead of reading the whole document yourself, you can ask the AI to surface gaps and issues.

For developers: download_file_as_text returns plain text. It works best on text-based formats (DOCX, TXT, MD). For files with complex formatting, some structure may be lost in the text conversion — your agent should handle this gracefully.

Step 4: Upload a revised version​

After incorporating feedback, you have a second draft ready. Upload it alongside the first.

Try this prompt:

Upload "launch-brief-v2.docx" to the "03 — Drafts & Work in Progress" folder.

Alternatively, if you want to update the existing file record rather than add a new one:

Update the file "launch-brief-v1.docx" with the revised version "launch-brief-v2.docx".

This uses update_file to replace the file content while preserving the same file record and ID.

When to use upload_file vs update_file:

  • Use upload_file when you want to keep both versions side by side (version comparison)
  • Use update_file when you want a clean replacement and don't need to preserve the old version

Step 5: Compare the two drafts​

With both versions available, compare them:

Download and compare "launch-brief-v1.docx" and "launch-brief-v2.docx". What changed between the two versions?

The AI will call download_file_as_text twice and then produce a diff summary. This is a lightweight alternative to tracked changes for teams who don't use collaborative editing features.

Step 6: Promote the approved document​

The brief has been approved by the client. Move it from the Drafts folder to the Final Deliverables folder.

Try this prompt:

Move "launch-brief-v2.docx" from "03 — Drafts & Work in Progress" to "04 — Final Deliverables" in the "Meridian Corp — Website Redesign" room.

The AI calls move_batch_items with the file ID and destination folder ID.

If you want to keep a copy in Drafts as a reference:

Copy "launch-brief-v2.docx" from "03 — Drafts & Work in Progress" to "04 — Final Deliverables".

This uses copy_batch_items instead, which leaves the original in place.

Key distinction:

  • move_batch_items — removes from source, places in destination
  • copy_batch_items — original stays in source, a copy is placed in destination

Step 7: Clean up old drafts​

Now that the approved version is in Final Deliverables, delete the first draft to keep things tidy.

Try this prompt:

Delete "launch-brief-v1.docx" from the "03 — Drafts & Work in Progress" folder.

The AI calls delete_file. Before proceeding, the AI might ask you to confirm since deletion is irreversible. Always confirm you have the right file before saying yes.

Final verification:

Show me the contents of both the "03 — Drafts & Work in Progress" and "04 — Final Deliverables" folders.

You should see only launch-brief-v2.docx in Final Deliverables and an empty (or cleaned-up) Drafts folder.