# OneNote MCP for AI Agents AI Agent Connect

> OneNote MCP connects your AI agent to your Microsoft OneNote notebooks. It lets your agent search through thousands of pages, read specific content from nested sections, and write new notes or summaries directly into your existing pages. It turns your static notes into a searchable, interactive knowledge base for your AI.

## Overview
- **Category:** industry-titans
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_v0VuD1Lg72idPZ9fSLCb2mmSOd0IthUQyOzaINkT/ai-agent-connect
- **Tags:** digital-notebook, content-management, information-retrieval, document-organization, personal-knowledge-base

## Description

You spend all day putting information into OneNote because it's where your thoughts live, but finding that one specific detail from three months ago feels like a chore. You have to click through nested sections and scroll past old headers just to find a single bullet point. This Connector changes that by giving your AI agent a direct line into your notebooks. Instead of you hunting for info, you can just ask your agent to find the 'Project Alpha specs' or 'Meeting notes from last Tuesday.' It scans your entire structure, pulls the raw text from the right page, and summarizes it for you right in the chat. You can even have the agent write back to your notes. If you're summarizing a meeting, the agent can take those notes and append them to your active work notebook without you having to switch tabs or copy-paste anything. It makes your personal knowledge base feel like a living database rather than a graveyard of old pages. When you use the Vinkius marketplace to connect this, you're basically giving your AI a memory that spans your entire professional and personal history. It's about moving from 'where did I put that?' to 'tell me what that says.' Whether you're managing a complex project or building a personal research library, this connection lets you treat your notes as a dynamic source of truth that your agent can actually interact with.

## Tools

### list_notebooks
List all your Microsoft OneNote notebooks so your agent knows where to start. This helps your AI understand the main categories of your notes.

### get_notebook
Get the specific properties and permissions for a single notebook. This lets your agent see the details of a specific container.

### list_pages
Show the titles and IDs of pages within a section without loading the heavy content. This lets your agent see a list of pages without wasting time.

### get_page_content
Pull the actual text and HTML from a specific page for your agent to read. This lets your agent see the full content of a page when you need details.

### search_pages
Search for keywords across all notebooks to find specific information quickly. This is perfect for finding a single fact buried in a large archive.

### list_sections
List the specific folders where your actual pages are stored. This allows your agent to drill down into your specific note categories.

### list_section_groups
Show the high-level folders inside a notebook for better navigation. This helps your agent see how you group your sections.

## Prompt Examples

**Prompt:** 
```
Find my notes about the 2023 marketing budget.
```

**Response:** 
```
### Marketing Budget 2023
I found 3 relevant notes in your **'Finance'** notebook. Here is a summary of the key points:

*   **Q1 Allocation:** $50,000 for social media ads.
*   **Q2 Strategy:** Shift toward influencer partnerships.
*   **Total Budget:** $200,000 confirmed by the executive team.

Would you like me to pull the full text from the 'Q1 Allocation' page?
```

**Prompt:** 
```
What were the main takeaways from the 'Project Phoenix' meeting?
```

**Response:** 
```
### Project Phoenix Meeting Summary
Based on the notes from **January 12th**, here are the main takeaways:

**Key Decisions:**
*   ✅ Approved the new UI prototype.
*   ✅ Moved the launch date to March 15th.

**Action Items:**
*   [ ] Backend team to fix the login bug.
*   [ ] Marketing to draft the press release.

*I can append these action items to your 'Project Phoenix' task list if you'd like.*
```

**Prompt:** 
```
List my notebooks so I can see what I have.
```

**Response:** 
```
Here are the notebooks currently in your OneNote account:

| Notebook Name | Status |
| :--- | :--- |
| **Personal Notes** | Active |
| **Work Reference Base** | Active |
| **Marketing Meetings 24** | Active |
| **Drafts** | Active |

Which one would you like to explore further?
```

## Capabilities

### Search every page in your notebooks
Your agent finds specific keywords across all your notebooks instantly.

### Pull raw text from specific pages
The agent reads the full content of a page to provide detailed summaries.

### Map your notebook hierarchy
Your agent sees the structure of your notebooks, section groups, and sections.

### Read specific page metadata
The agent identifies page titles and IDs without loading heavy content.

### Append new notes to sections
Your agent adds summaries or new thoughts directly into your existing notebooks.

## Use Cases

### Retrieving old project specs
A researcher needs a specific quote from a year ago. They ask the agent to search for 'climate data 2022' and it finds the page via search_pages.

### Mapping out a new project structure
A project manager wants to see the structure of a new project. They ask the agent to list the sections in the 'Project Alpha' notebook using list_sections.

### Summarizing meeting notes
An assistant needs to summarize a long meeting. They ask the agent to read the notes from 'Meeting Jan 5' using get_page_content and provide a summary.

### Building a personal knowledge base
A writer wants to see what's in their 'Work' notebook. The agent uses list_notebooks to show the main containers first before diving into specific pages.

## Benefits

- Stop manual searching by using search_pages to find keywords across your entire OneNote library in seconds.
- Extract precise data from deep archives using get_page_content to pull raw text without clicking through folders.
- Map your entire workspace with list_notebooks and list_section_groups so your agent understands your organizational logic.
- Save time on data entry by having your agent summarize thoughts and put them into your notes directly.
- Connect your research to your chat by letting your agent pull citations from your personal knowledge base.
- Keep your workspace clean by letting your agent handle the navigation of complex nested section groups.

## How It Works

The bottom line is your AI gets a direct window into your private OneNote archives to find and use your own data.

1. Connect your Microsoft Graph Access Token to the Connector
2. Type a natural command like 'Find my notes on Project Alpha' in your AI client
3. Get a summary or a list of matches from your notebooks instantly

## Frequently Asked Questions

**Can the OneNote MCP search my private notes?**
Yes. It connects directly to your notebooks, allowing your AI agent to search through your private notes and find specific information as if it were reading them with you.

**How does the OneNote MCP handle nested sections?**
It maps out your entire organizational logic. The agent can navigate through notebooks, section groups, and individual sections to find exactly where your data is stored.

**Is the OneNote MCP good for a personal knowledge base?**
It is perfect for that. It turns your personal notes into a searchable database, allowing you to ask your agent for summaries, citations, or facts from your own history.

**Does the OneNote MCP support enterprise notebooks?**
Yes, it works with both personal and enterprise notebooks, provided you have the correct Microsoft Graph access tokens to reach them.

**How do I get my AI to read a specific page?**
You can simply ask your agent to find a page by its title or a keyword. Once it identifies the page, it can pull the full text to answer your questions or provide a summary.

**Can the OneNote MCP help with research?**
Absolutely. Researchers can use it to scan thousands of pages of notes for obscure citations or specific data points, saving hours of manual scrolling.

**Can the integration delete entire extensive notebooks or important local sections?**
No. The integration exclusively binds heavily to Reading methods (list, search, get) mapped safely alongside minimal Write interactions specifically scoped to Appending fresh notes. Destructive end-points are intrinsically restricted protecting vital long-term data persistently.

**Does it parse images contained inside OneNote native pages automatically?**
Currently, fetching Page contents operates exclusively translating returned DOM hierarchies returning raw string-based text bounds flawlessly. Purely visual bitmaps or native handwritten encodings skip the explicit parsing stream avoiding LLM hallucinations fundamentally.

**Why does OneNote demand a generic Graph Access Token explicitly here?**
Microsoft explicitly deprecated discrete independent silo APIs unifying all native core tenant structures explicitly over the single Microsoft Graph perimeter gateway. OneNote resources are canonically navigated traversing this exact Graph endpoint hierarchy natively.