How to Connect Claude Desktop to Your Document Workspace
Your tracked documents already know who opened them, what got read, and which links are live. The only cost of that knowledge is the trip: open the dashboard, find the document, click the tab. Most people stop making the trip.
Connect Claude to Apdf over MCP and the trip disappears. Your workspace becomes something you ask: who read the July update? send Elif the pricing guide. kill Dana's link. One sign-in, then it's conversation.
Connect — one URL, one sign-in
In Claude Desktop, go to Settings → Connectors → add a custom connector and paste the endpoint:
https://apdf.io/mcp/mainAn OAuth window opens; sign in with your Apdf account and you're connected — no API keys, no config files. The connection is scoped to your workspace: Claude sees what you see, nothing more. In Claude Code the same thing is a one-liner:
claude mcp add --transport http apdf https://apdf.io/mcp/mainStart with inventory
First prompt: find out what Claude can see.
Under the hood that's the list-documents tool, and
every entry carries the essentials the follow-up questions build on:
{
"doc_id": "6eb41-45fe5-bd354",
"viewer_url": "https://docs.apdf.io/studio/6eb41-45fe5-bd354",
"file_url": "https://apdf-files.s3.eu-central-1.amazonaws.com/docs/202607/b2388c226a61cf555e433.pdf",
"name": "Q3 Board Pack",
"file_size": 77952379,
"file_pages": 12,
"is_public": false,
"created_at": "2026-07-23T08:23:00.000000Z",
"archived_at": null
}Ask the questions you never open the dashboard for
{
"session_id": "wGVOCQpiPzWBN4hz",
"viewer_id": "z9rRE2e40yJ2249K",
"recipient": {
"name": "Sandra Okafor",
"email": "sandra@okaforcap.example"
},
"device_type": "desktop",
"browser": "Chrome",
"os": "OS X",
"referer": null,
"utm_source": null,
"created_at": "2026-07-22 14:39:30.000",
"pages_viewed": 4,
"total_duration_ms": 231000,
"completion_pct": 100,
"prints": 0,
"downloads": 0
}Sessions, aggregate analytics, annotations, form submissions, automation runs — each has its own tool, which means each is one question away. The agent picks the right one; you never learn the names.
Then stop asking and start doing
The connection isn't read-only. Sharing, revoking, archiving — the write side works the same way:
{
"token": "aLo58OJu",
"name": "Elif Demir",
"email": "elif@demirstudio.example",
"url": "https://docs.apdf.io/studio/688b0-07ba8-42bc4?v=aLo58OJu",
"is_active": true,
"last_viewed_at": null,
"created_at": "2026-07-23T09:17:50.000000Z"
}The same toolbelt carries deactivating and restoring links, flipping documents private, archiving — and the full set of PDF operations (merge, split, OCR, compress, create), so the document you're discussing can also be transformed in the same conversation.
Where to go from here
Connected? These are the conversations to have next.
Related tutorials
Build a Morning Briefing Agent for Your Documents
A digest lists what happened; a briefing arrives with the thinking done. One standing prompt over the MCP-connected workspace and Claude sweeps every document's sessions on a headless cron — call-first with evidence and a suggested opener, the next tier in one line each, and the silence flagged.
Ask Claude Which Deals Are Hot
The morning question isn't about one document — it's 'across everything, who should I call first?' With the workspace connected over MCP, the agent sweeps every document's sessions, ranks re-opens over deep reads over skims, and answers with evidence: the Arslans came back for the price page.
Mint PDF Tracking Links from ChatGPT
The share decision happens mid-conversation, not in a dashboard tab. The MCP server speaks to any client — including ChatGPT — so 'share the pricing guide with Elif' mints her tracked link inside the draft you were writing, and the same chat checks who read and kills a leaver's old link.