Last updated:
Wave AI runs a hosted MCP server at mcp.wave.co that lets Claude, ChatGPT, Claude Code, Cursor and other MCP clients search your recorded meetings, read speaker-labeled transcripts and work with action items. Wave is listed as Wave AI Note Taker in the Claude connector directory and the ChatGPT app directory, so connecting takes one click and a sign-in.
The Model Context Protocol (MCP) is an open standard that lets AI assistants call external tools. Wave's MCP server exposes your Wave recordings as tools, so Claude or ChatGPT can search your meetings and read transcripts, summaries and action items, instead of you copying and pasting them.
Open Wave AI Note Taker in the Claude connector directory, select Connect and sign in to the Wave account you record with. It then works in Claude on the web, desktop and mobile, and in Claude Code.
Wave for ClaudeInstall Wave AI Note Taker from ChatGPT's directory and sign in to Wave. ChatGPT can then use your meetings in regular chats, company knowledge and Deep Research, with no Developer Mode.
Wave for ChatGPTCursor, Codex, Grok and any client that speaks MCP can connect to https://mcp.wave.co with OAuth 2.0 and PKCE, or with a manual wave_mcp_ token you create in Wave's settings.
Wave developer portalClaude and ChatGPT can't see your conversations on their own, and copy-pasting transcripts into a chat doesn't scale. Wave's MCP server gives your assistant direct, permissioned access to your meeting history.
Most note takers are a destination: you go to them, search, copy things out. MCP inverts that, so your notes come to wherever you already work with AI.
Drafting a follow-up email, a project plan or a status update means reassembling context by hand. With Wave connected over MCP, your assistant drafts from the transcript itself.
Read tools include search_sessions, list_sessions, get_session, get_transcript, get_action_items, list_people and get_person_sessions. Write tools include update_action_items, update_session, the folder tools and share_session.
search_sessions runs semantic search across your recordings, so your assistant finds 'the call where the client pushed back on timeline' without you remembering which meeting it was.
When your assistant needs verbatim detail, such as exact quotes or who said what, get_transcript returns the full speaker-labeled transcript of any session.
get_action_items and update_action_items let an assistant read and update the action items from your meetings, so 'what's still open from this week's calls?' gets a real answer.
Six prompt templates come with the server: weekly-recap, prep-for-meeting, follow-up-email, action-items-to-tasks, what-did-person-say and decisions-log.
OAuth 2.0 with PKCE: you approve access explicitly, scoped to your own account, and can disconnect anytime. Importing recordings needs a separate opt-in permission. Wave never trains AI models on your recordings.
Wave AI Note Taker is live in the Claude connector directory and the ChatGPT app directory. Connect in one click and ask either assistant about any meeting you have recorded.
You can now connect Wave to Claude. Once it’s linked, Claude can search, read, and summarize any meeting you’ve recorded — so instead of opening the app to hunt for a transcript, you just ask. Setup takes about a minute and no code.
The complete guide to recording with Wave — iPhone, Android, Apple Watch, Wear OS, Desktop, Chrome Extension, Meeting Bots, Phone Calls, Web App, and Import. Every way to capture, transcribe, and summarize your conversations.
Wave AI is an AI meeting assistant for Zoom, Google Meet, Teams and in-person meetings: it records, writes speaker-labeled transcripts and summaries, and answers questions. Free to start.
Wave AI is an AI note taker app for iPhone, Android, Mac, Windows and the web. Record meetings, calls and lectures; get transcripts, summaries and action items. Free for 30 minutes a month.
Stop taking meeting notes. Wave AI records Zoom, Meet, Teams or in-person meetings and writes the notes for you: decisions, action items and who said what, minutes after. Free to start.

Product
Use Cases
Compare
All rights reserved
Made with love in New York City