MCP Setup Guide: Connect Claude to Your Whole Business
Most AI setups die in the same spot: the assistant is smart, but it cannot touch your actual work. This guide goes straight to the wiring — which client to pick, which servers to connect first, and the one safety rule that keeps a connected AI from becoming a liability.
What This Guide Covers
A plain-language MCP setup guide for connecting Claude to your real tools — no engineering background required. You will get the client choice (Claude Desktop vs. Claude Code), the order to connect Drive, Notion, Gmail, and Calendar, why read-only access should come first, and a sample first task to run once everything is live.
Why Your AI Still Can't See Your Actual Work
The gap is not intelligence. It is access.
A founder running a lean team usually hits the same wall a few weeks into using Claude or ChatGPT seriously: the model is good at reasoning about a problem once you describe it, but it cannot see the problem sitting in your own systems. It cannot open the Drive folder with last quarter's numbers. It cannot check whether a Notion task is actually done. It cannot read the client email that is sitting unanswered right now. Every session starts from zero context, and every answer is only as good as what you manually copy in.
Model Context Protocol (MCP) closes that gap. It is the open standard, released by Anthropic in November 2024, that lets an AI client connect to outside tools and data through one shared interface instead of a custom integration for every pairing.
- Pick a client that supports MCP (Claude Desktop or Claude Code).
- Connect your first 1–2 servers — start with the tool you open most every day.
- Give it read-only access and verify results are accurate before trusting it to act.
- Add write access one tool at a time — only after confirming it reads correctly.
Step 1: Pick Your Client
The client is the AI application MCP servers plug into. For most people, that means choosing between Claude Desktop and Claude Code.
The app you open to chat — the same window you already use for everyday questions. Adding MCP servers to it means every conversation you start can reach your connected tools without any extra setup per chat.
The terminal-based tool built for people working directly in a codebase or repository. It runs the same MCP connections but is built for tasks that involve files, git, and multi-step automation.
If your first goal is "let Claude check my Notion and my inbox," start with Claude Desktop. If your goal involves a codebase, a repository, or a publishing pipeline, Claude Code is the better starting point. Nothing stops you from running both once you are comfortable with either.
Step 2: Connect Drive, Notion, Gmail, and Calendar
Connect one tool at a time, starting with whichever one you check most often during a normal workday.
Notion (Easiest First Connection)
Notion is usually the easiest first connection. Here is how it works:
- Add the Notion MCP server using an integration token from your Notion settings.
- Explicitly share the specific pages or databases you want Claude to see — nothing is visible by default.
- You choose what to expose, one page or database at a time.
- Go to Notion Settings → Integrations → New Integration, name it, and copy the token it generates.
- In Claude Desktop, go to Settings → Integrations and paste your Notion token to add the server.
- Then open each Notion page you want Claude to access and click Connect Integration in the top-right corner.
Gmail, Google Drive & Calendar (One OAuth Flow)
Gmail, Google Drive, and Calendar sit behind the same Google authorization flow, so setting up one makes the other two faster.
- Expect an OAuth consent screen rather than a simple API key.
- Sign in with your Google account and approve the specific scopes requested (read your inbox, read your Drive files, etc.).
- The client stores that authorization for future sessions — you only approve once.
- Start with read scopes only: "read inbox," "read Drive files," "read calendar events." Do not approve write scopes until you have confirmed read accuracy across several real tasks.
Resist the urge to connect everything on day one. Every connected server adds its tool definitions to the client's context, and a bloated tool list measurably degrades how well the model picks the right tool for a given request. Start with the two or three tools you actually open every day — not the full list of everything MCP can theoretically reach.
Step 3: Read-Only First, Actions Second
Every connection should start read-only — meaning Claude can see and summarize, but cannot create, edit, delete, or send.
This is not caution for its own sake. It is how you learn what the connection actually returns before you trust it with a write action.
Test Prompts to Run During the Read-Only Phase
Watch what comes back. Does it pull the right rows? Does it miss anything? Does it hallucinate a detail that is not actually in the source? A read-only phase surfaces those gaps with zero downside, because nothing gets changed on the other end no matter what the model gets wrong.
Once the read-only answers are consistently accurate across several real tasks — and only then — move to granting write access on that one connection: sending a draft reply, updating a Notion status, creating a calendar event. Do this one tool at a time, not all at once.
The One Safety Rule Before Any Tool Gets Write Access
Never grant write access to a connection you have not first watched succeed at reading — more than once — on a real task.
This matters because of how the protocol itself is built. Anthropic's Model Context Protocol specification, revised on July 28, 2026, formalizes that tool-initiated actions go through an explicit consent path rather than running silently in the background. The client is expected to surface what a tool is about to do before it does it.
That design assumes a human is actually watching the first few times a new connection acts — not rubber-stamping every prompt on autopilot. Skip the read-only phase, and you are the one breaking that assumption, not the protocol.
- User consent before a connected tool takes any action is now formally required in the spec.
- Streamable HTTP is now the standard transport for remote MCP servers, requiring an explicit method header on every request.
- The protocol under the hood is actively hardening its security model as adoption grows — which is exactly why starting cautiously on your end still matters.
Your First Task Once You're Connected
The fastest way to know the setup actually works is to give it one real task — not a demo prompt.
The 5-Step Read-Only Test Workflow
Run this exact sequence when you first connect Notion. If it completes cleanly, your setup is working.
In Harshal's own workflow, once Notion, GitHub, and a deployment connection were live, the first real task was: "check the content database for anything marked ready, write it up, and open a pull request." Claude found the record, read the details, drafted the file, and pushed a branch — the same kind of pipeline that produced the original blog post.
Key Takeaways
- Pick a client first. Claude Desktop for everyday chat tasks (Notion, Gmail, Calendar). Claude Code for codebases, repos, and automation pipelines.
- Notion is usually the easiest first server. An integration token plus explicit page or database sharing. Nothing is visible by default.
- Gmail, Drive, and Calendar share one Google OAuth flow. Setting up one speeds up the other two.
- Cap your initial install at 3–5 servers. More than that visibly degrades how well Claude picks the right tool for a task.
- Every connection should start read-only. Only grant write access after watching a connection succeed at reading — more than once — on a real task.
- The July 28, 2026 MCP spec update formalizes user consent before a tool acts and pushes Streamable HTTP as the standard transport for remote servers.
- The clearest sign a setup works is a real task completed end to end — not a demo question answered once.
Quick Reference: Client & Scope Cheat Sheet
| Tool | Auth Method | Read-Only Scope | Write Scope (After Verification) |
|---|---|---|---|
| Notion | Integration Token | Read pages, databases | Update properties, create pages |
| Gmail | Google OAuth | Read inbox, read labels | Send emails, create drafts |
| Google Drive | Google OAuth | Read files, list folders | Create, edit, upload files |
| Google Calendar | Google OAuth | Read events, read attendees | Create events, invite attendees |
| GitHub | Personal Access Token | Read repos, read issues | Push branches, open PRs |
Frequently Asked Questions
Do I need to know how to code to set up MCP?
No. Notion and most one-click integrations only need a token and a few clicks. Gmail, Drive, and Calendar involve an OAuth consent screen — not custom code. Claude Code setups can involve editing a config file directly, but Claude Desktop setups are mostly point-and-click once you have the token or OAuth completed.
Which should I connect first — Notion or Gmail?
Whichever one you check most during a normal workday. Notion is usually faster to set up since it only needs an integration token and explicit page sharing. Gmail requires the Google OAuth flow but then unlocks Drive and Calendar at the same time.
Is it safe to give Claude access to my inbox and Drive?
It is safer once you start read-only and confirm the results are accurate before granting any write access. The 2026 MCP specification also requires explicit consent before a connected tool takes any action — but the practical safety comes from you watching the read-only phase before enabling writes.
How many MCP servers should I connect at once?
Three to five to start. Every server adds its tool definitions to Claude's active context. Too many and the model starts picking the wrong tool for a given request — it becomes slower and less accurate. Add more only after your initial connections are running smoothly on real tasks.
What is the difference between Claude Desktop and Claude Code for MCP?
Claude Desktop is the everyday chat app — best for looking things up in Notion, Gmail, or Calendar and having a conversational back-and-forth. Claude Code is a terminal tool built for file, git, and multi-step tasks like the one that drafted and pushed the original blog post. Both support MCP connections; the difference is what kind of work fits each interface.
This guide is based on the blog post MCP Setup Guide: Connect Claude to Your Whole Business. Read it for the full context, including real examples from Harshal's own workflow.
Harshal Saraf
Creative Director + AI Workflow Consultant
Harshal Saraf is a Creative Director and AI Workflow Consultant based in Indore, India. Under his practice ByHarshal, he sets up AI workflows for founders, agencies, and brands across India — including connected MCP pipelines that automate content, research, and client communication end to end.