Granola MCP. Stop reading notes. Start getting answers.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Granola.ai MCP Server manages your meeting notes and conversation memory. It lets your agent list, search, and retrieve full transcripts, summaries, and action items from any meeting document.
You can track key decisions and follow up on tasks directly from the chat, without manual note-taking.
What your AI agents can do
Get action items
Extracts specific tasks and owners identified from a given meeting document.
Get content
Retrieves the complete, structured text content of a specific meeting document.
Get documents batch
Fetches multiple meeting documents using a list of IDs in one request.
List documents by date range or browse the entire workspace hierarchy to locate specific meeting folders and files.
Run a full-text search across every document in the workspace to pinpoint discussions about a specific topic.
Ask the agent for an AI-generated overview of a document, pulling out the main takeaways and key points.
Automatically extract all specific action items and assignees from a meeting document to build a task list.
Retrieve a complete, speaker-detected transcript of a meeting, including timestamps and speaker names.
Get metadata (like date, owner, and type) for a single meeting document.
Cross-reference a meeting ID with calendar data to retrieve the list of participants.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
Granola MCP Server: 12 Tools for Meeting Intelligence
Use these 12 tools to manage every aspect of your meeting data, from listing documents to extracting specific tasks and full transcripts.
019d75aaget action items
Extracts specific tasks and owners identified from a given meeting document.
019d75aaget content
Retrieves the complete, structured text content of a specific meeting document.
019d75aaget documents batch
Fetches multiple meeting documents using a list of IDs in one request.
019d75aaget metadata
Retrieves key data points, like dates and authors, for a single meeting document.
019d75aaget participants
Gets a list of people who attended a specific meeting session.
019d75aaget summary
Generates a concise, AI-written summary of a meeting document.
019d75aaget transcript
Retrieves the full, speaker-detected transcript of a meeting with timestamps.
019d75aalist by date
Lists all meeting documents that fall within a specified date range.
019d75aalist documents
Lists all meeting documents in the workspace, supporting pagination.
019d75aalist folders
Lists all categorized folders (directories) in the workspace hierarchy.
019d75aalist recent
Lists the 20 most recently created meeting documents.
019d75aasearch documents
Performs a full-text search across every document in the workspace.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Granola, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
Granola.ai MCP Server manages your meeting notes and conversation memory. Your agent lets you list, search, and pull full transcripts, summaries, and action items from any meeting document. You can track key decisions and follow up on tasks right from the chat, without having to take notes yourself.
Find All Meeting Records
- You can list all meeting documents in the workspace, or you can list the 20 most recent ones. You can list all documents by checking the folder structure or by checking documents within a specific date range. You can also run a full-text search across every document to pinpoint discussions about a specific topic.
- You can check the high-level folder labels in the workspace hierarchy by listing all folders.
Search Across All Meeting Notes
- Run a full-text search across every document in the workspace to find specific discussions.
Generate Meeting Summaries
- Ask your agent for a concise, AI-written overview of a document, pulling out the main takeaways and key points.
Identify Follow-up Tasks
- Your agent automatically extracts all specific action items and assignees from a meeting document, building a task list for you.
Get Full Transcripts
- You can retrieve a complete, speaker-detected transcript of a meeting, including timestamps and speaker names.
Check Document Details
- You can check key data points like dates, authors, and types for a single meeting document by getting its metadata.
List Meeting Attendees
- Your agent retrieves the list of people who attended a specific meeting session using the meeting ID.
Retrieve Content and Batch Data
- You can get the full, structured text content of a specific meeting document.
- You can fetch multiple meeting documents using a list of IDs in one request.
How Granola MCP Works
- 1 1. Subscribe to this server and enter your Granola API Key (found in your Granola Settings).
- 2 2. Your AI client sends a natural language request (e.g., 'What were the action items from the Q2 review?').
- 3 3. The server calls the appropriate tool (like
get_action_items), and your agent gets the structured data back instantly.
The bottom line is that your agent manages meeting notes and searchable memory without you having to manually copy, paste, or switch between tools.
Who Is Granola MCP For?
Anyone drowning in meeting notes and transcripts. This is for the PM who needs to audit weeks of discussions, the Manager who tracks action items from back-to-back calls, or the EA who needs to organize and find participant data across dozens of folders. It stops you from wasting time reading notes you already read.
Auditing collaborative meeting documents and searching across conversation history using natural language queries.
Summarizing back-to-back meetings and tracking action items without taking manual notes.
Organizing meeting folders and retrieving participant metadata through the chat interface.
Verifying organizational knowledge and finding meeting results in real-time without manual data aggregation.
What Changes When You Connect
- Automatically track follow-up tasks. Instead of manually reading minutes to find 'Who needs to do what,' use
get_action_itemsto isolate every specific task and owner from any meeting record. - Cut through the noise with summaries. Don't read 50 pages of meeting transcripts. Use
get_summaryto get a synthesized AI overview that hits the key decisions and takeaways immediately. - Search the whole history. Need to know what was said about 'Q3 budget' three months ago? The
search_documentstool runs full-text detection across your entire knowledge base. - Audit meeting details. Use
get_metadataandget_participantstogether to confirm who was in the room and when the document was finalized, giving you full context. - Handle bulk data efficiently. When you need to compare 10 different meetings, use
get_documents_batchto fetch all their content in a single, secure request. - Get the full story. Need the exact dialogue?
get_transcriptpulls the raw, speaker-detected dialogue, including timestamps, so you can audit the conversation exactly.
Real-World Use Cases
Need to report on a specific project phase.
A PM needs to summarize all discussions around 'Project Zenith' from the last quarter. They ask their agent to use list_by_date to find all relevant meetings, then run get_summary on each one, and finally compile the key decisions into a single report. The problem is solved in one flow.
Follow up on action items after a large team meeting.
A Team Lead runs a meeting and needs to ensure follow-up tasks are assigned. They ask their agent to use get_action_items on the transcript. The agent returns a clean list: 'John owns X, Sarah owns Y, due by Z.' No manual cross-referencing is needed.
Find a quote from a conversation months ago.
An Ops Specialist needs to verify a detail mentioned in a meeting from last year. They use search_documents with keywords. The tool finds the relevant document, and the user can then ask for the get_content of that specific passage.
Compare multiple meeting outcomes quickly.
An Executive Assistant needs to compare the scope of the Q1 and Q2 strategy meetings. They use get_documents_batch to pull the content of both documents simultaneously, allowing them to analyze cross-meeting dependencies in one view.
The Tradeoffs
Treating the server like a search bar
Asking the agent to simply 'search for all summaries.' The agent might just return a list of links, forcing you to click each one to read the summary.
→
Always specify the action. Instead of searching, tell the agent to search_documents and then explicitly ask it to run get_summary on the top three results. This forces the agent to process the data, not just list the source.
Asking for everything at once
Prompting: 'Give me the summary, the transcripts, the action items, and the participant list for the whole quarter.' The agent will likely fail due to scope overload.
→
Break it down. First, use list_by_date to narrow the scope. Then, use get_documents_batch to pull the relevant IDs. Finally, ask the agent to run get_summary and get_action_items on that batch of IDs.
Assuming one tool covers all data
Relying only on get_content to find the next steps. The raw content is too dense and will bury the action items in paragraphs of discussion.
→
Always run get_action_items alongside get_content. The specialized tool isolates the tasks, giving you structured, actionable data separated from the discussion text.
When It Fits, When It Doesn't
Use this server if your job requires auditing, comparing, or synthesizing information from dozens of meeting documents. It's for the knowledge worker who needs to act on what was said, not just read it.
Don't use it if your only goal is to find a single document name. Use the list_documents tool first. If you just need a basic list of files, list_documents is faster. If you need to find a specific discussion topic across many years of records, then the power of search_documents is necessary. If you're dealing with structured data like calendars, check get_participants first. The key is knowing if you need a list (browse/search) or an extraction (summary/actions).
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Granola. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 12 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Sifting through meeting minutes and transcripts is a time sink.
Today, you manually open the calendar, find the meeting name, open the associated document, and scroll through pages of notes. If you need a specific detail—like who was supposed to update the landing page—you have to read through the entire document, cross-referencing notes and transcripts. It's a full-time job just to find three action items.
With this Granola MCP Server, you ask your agent directly: 'What were the action items from the Product Design Workshop?' The server runs `get_action_items` and returns a clean, structured list instantly. You get the data, not the document.
Granola MCP Server: Extract Action Items and Summaries
You used to have to copy-paste the raw transcript into a separate AI tool, then manually prompt it to summarize, and then copy the summary into a project management board. It's a three-step, multi-tool process that always failed because you lost context.
Now, your agent runs `get_summary` and `get_action_items` independently. The system processes the meeting data, and you get two separate, structured outputs: one concise overview, and one clean list of tasks. It works every time.
Common Questions About Granola MCP
How do I use the `get_action_items` tool? +
You specify the meeting document ID or name and ask the agent to run get_action_items. It returns a structured list of tasks, including the assignee and the goal, separate from the discussion text.
Can I search across all meeting notes using `search_documents`? +
Yes, search_documents performs full-text detection across every document in the workspace. You just need to provide the keywords or phrases you're looking for.
What is the difference between `get_content` and `get_summary`? +
The get_content tool retrieves the full, structured text of the document, including annotations. The get_summary tool runs an AI model to create a concise, high-level overview.
How many meetings can I analyze with `get_documents_batch`? +
You can fetch multiple documents by providing their IDs in a single request. This is better than running 10 individual API calls, especially when analyzing complex cross-meeting dependencies.
Does `get_transcript` include speaker names? +
Yes, get_transcript retrieves the full transcript with speaker detection. It parses the speech data locally, giving you chronological metadata for auditing purposes.
How do I list all meeting documents using `list_documents`? +
It lists every meeting document in your workspace. This function supports pagination, so you can pull all records, even if there are thousands of meetings.
What does `get_metadata` provide about a meeting document? +
It returns specific data points about a single document. You get details like the creation date, last modified time, and the document's unique ID.
Can I pull multiple documents and their summaries using `get_documents_batch`? +
Yes, you can fetch multiple documents by their IDs in one request. This is ideal for analyzing cross-meeting dependencies quickly.
Can my agent retrieve the AI summary of a specific Granola meeting? +
Yes. Use the 'get_summary' tool. The agent will fetch the native generative results, reducing the bulk meeting content into synthesized blocks and key takeaways flawlessly.
How do I extract action items from recorded notes via chat? +
Use the 'get_action_items' tool. Your agent will isolate targeted steps inferred from the meeting intent, enabling you to automate your post-meeting follow-up workflows natively.
Can I search for meetings from a specific date range through the agent? +
Absolutely. Use the 'list_by_date' tool. Provide the start and end dates in ISO 8601 format. The agent will trigger temporal scoping routines to expose specific documents mapped historically.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
SoundCloud
Manage your SoundCloud account — audit tracks, playlists, and followers via AI.
Nylas
Equip your AI agent to manage emails, calendars, and contacts across all providers (Gmail, Outlook) through a single unified interaction.
EasySendy
Send high-volume email campaigns through multiple delivery servers with list hygiene tools that protect your sender score.
You might also like
Copysmith
Generate marketing copy, product descriptions, and ad variations at scale with AI trained on high-performing content.
Clickatell
Deliver transactional and marketing messages over SMS, WhatsApp, and chat channels with global reach and high deliverability.
Brightcove
Manage your video library via Brightcove Video Cloud — list videos, manage playlists, and monitor performance directly from any AI agent.