Model Context Protocol is an open standard that lets AI assistants connect to external tools. Memo AI ships an official MCP server, so any compatible client can work with your transcripts.

The key is scoped to a single workspace and shown only once at creation. You can revoke it anytime, and transcripts stay encrypted on Memo AI's side.

Model Context Protocol

Connect your meetings to Claude, ChatGPT and any AI assistant

Q: How do I connect meeting transcripts to Claude or ChatGPT?

Create an API key under Settings → API & MCP, paste the MCP server config into Claude, ChatGPT or Cursor (for Claude Code it's one command), and ask in plain language.

Q: Which assistants are supported?

Any MCP-compatible client: Claude Desktop, Claude Code, Cursor, ChatGPT (connectors on Plus/Pro/Team), VS Code, Windsurf and more. They all connect the same way — a config or one command.

Q: Do I need to know how to code?

No. For MCP you paste a ready-made config once, then talk to your assistant in plain language. If you do need a code integration, the same key works over the REST API.

Q: Which plan do I need?

API & MCP access is included in the Pro and Expert plans. Create your key under Settings → API & MCP.

Q: Do I need to install anything?

Almost nothing. The MCP server runs via npx — you only need Node.js 18+ installed. You paste the config into Claude, Cursor or ChatGPT once; nothing else to install.

The official MCP server opens your transcripts to your assistant

Create API key Documentation

{
  "mcpServers": {
    "memo-ai": {
      "command": "npx",
      "args": ["-y", "memoai-mcp-server"],
      "env": { "MEMO_API_KEY": "mk_live_…" }
    }
  }
}

What tasks came out of yesterday's meeting?

memo_get_transcription

From the May 28 meeting — 4 tasks: Anna — Friday release, Sergey — UI fixes, Maria — landing copy, Alex — ad budget.

Who owns the CRM integration?

memo_list_transcriptions

Discussed on two calls: owner — Sergey, due June 10. Details in the May 22 recording, 11:40.

One config, any client

Works with the assistants you already use

Ask anything about your meetings

Ask about any meeting

“What did we decide on pricing last week?” Your assistant finds the conversation and answers with a link to the exact moment — no re-listening

You

What did we decide on pricing last week?

memo_list_transcriptions → memo_get_transcription

On the Acme call (May 20) you agreed: base plan $29, enterprise on request. Alex locked the decision at 14:32.

Checklist analytics

Speech feedback

Search the archive

Export to notes

Recurring routines

You

What did we decide on pricing last week?

memo_list_transcriptions → memo_get_transcription

On the Acme call (May 20) you agreed: base plan $29, enterprise on request. Alex locked the decision at 14:32.

You

Did the reps cover every onboarding step on these calls?

memo_list_transcriptions

Checked 8 calls. Intro — 8/8, product demo — 7/8, pricing — 5/8. The “next contact” step was skipped in 3 conversations.

You

Review my pitch: structure, clarity, filler words.

memo_get_transcription

Structure is clear, but the intro runs long (2:40). Filler words: “like” ×12, “you know” ×9. Strong close with a call to action at 18:05.

You

Which calls had clients complaining about onboarding?

memo_list_transcriptions

Found 4: Acme (May 12), Globex (May 18), Initech (May 24), Umbrella (May 29). Shared pain — an unclear first step.

You

Save this week's meetings to ~/notes/meetings in markdown.

memo_bulk_export → ~/notes/meetings

Saved 7 .md files with frontmatter. Ready to open in Obsidian.

You

Summarize all client calls this week and save them to ~/team/weekly.

list → get ×N → bulk_export

Processed 12 calls, summarized each, saved 12 files to ~/team/weekly. Ready to share with the team.

Export all your meetings into Obsidian

Export meetings to Markdown — they drop straight into your Obsidian. Local, and ready for its AI

Markdown exportOffline accessUp to 100 meetings at onceStored locallyReady for Obsidian AI

Export to Obsidian

Obsidian

Memo Vault

Meetings

May 2026

Acme — discovery call.md
Weekly sync.md
1:1 with Jordan.md
Product roadmap Q3.md

Acme — discovery call

2026-05-20

Alex, Jordan

Pricing, Onboarding

Summary

–Agreed on the base plan and enterprise terms.
–Decided to simplify the first onboarding step.

Connect in two minutes

One key, and your assistant is ready to work with your meetings

Create an API key

In the web app, open Settings → API & MCP and click “Create key”. It starts with mk_live_ and is shown only once

Connect your assistant

Paste the ready-made config into Claude, ChatGPT or Cursor. For Claude Code it's a single terminal command — no file editing

Ask in plain language

“Find my meetings this week and collect action items.” Your assistant reaches out to Memo AI and answers from your real transcripts

What your assistant gets

Tools the assistant calls on its own — you just describe the task

Finds the right meetings

Search by title, summary, speakers and topics — with filters for date, project and language

›memo_list_transcriptions

Reads the full meeting

Complete text with speakers, timestamps and ready AI reports for a single conversation

›memo_get_transcription

Saves to disk

Exports a meeting as .md or .txt straight to your folder — keeping the chat light and saving tokens

›memo_export_transcription

Exports in bulk

Up to 100 transcripts in a single request — export an entire project at once

›memo_bulk_export

Sees your projects

Lists workspace projects so you can filter meetings by the exact track you need

›memo_list_projects

Why an MCP server beats a plain AI chat

Answers linked to the moment

The assistant relies on the real timestamped transcript, not a paraphrase — jump straight to the right second of the conversation

Context from hundreds of meetings

Search runs across your whole archive at once — decisions and commitments surface from dozens of conversations in one request

Export instead of a wall of text

Long meetings are saved to disk as a file rather than dumped into the chat — the dialog stays light and tokens don't burn

One key, no code

The same mk_live_ works in Claude, ChatGPT and Cursor. No server to run, no integration to write — paste the config and go

lines of code

Paste one config — the assistant calls the right tools itself

100

meetings per request

Export an entire project in a single call

~2 min

to connect

From creating a key to your first answer from your meetings

key for everything

One API key for any MCP client and the REST API

Frequently asked questions

The short version on setup, access and security

Does Memo AI have an MCP server?

Yes. Memo AI provides an official MCP server for meeting transcripts — it works with Claude, ChatGPT, Cursor and any MCP-compatible client.

How do I connect meeting transcripts to Claude or ChatGPT?

What is MCP?

Which assistants are supported?

Do I need to know how to code?

Which plan do I need?

Is it secure?

Do I need to install anything?

Give your AI access to your meetings

Create a key, paste the config — and ask your assistant about any conversation within minutes

Create API key How to connect