Skip the install. Get this working in under 2 minutes.
Start a free trial on cloud.anythingmcp.com, add the Otter.ai in one click, then point your AI client (Claude, ChatGPT, Copilot or Cursor) at the generated MCP endpoint. No Docker, no git clone, zero engineering experience required.
Summary
Let your AI read your Otter.ai meeting notes: list recorded conversations, read Otter's summaries, action items, insights and transcripts, browse channels, and with your confirmation import a recording from a link. Needs an Otter Enterprise workspace.
Try asking
Example prompts for Otter.ai
Click any prompt to copy it. Paste into Claude, ChatGPT, Cursor, Gemini, Copilot or OpenClaw to run it against this connector.
Claude is AI and can make mistakes. Please double-check responses.
💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → Otter.ai, paste your credentials, mint an MCP API key and you're done. No Docker, no
git clone, no local server.
Otter.ai + ChatGPT
Let your AI read your Otter.ai meeting notes: list recorded conversations, read Otter's summaries, action items, insights and transcripts, browse channels, and with your confirmation import a recording from a link. Needs an Otter Enterprise workspace.
Prerequisites
See the full setup instructions baked into the connector (visible in the in-app store when you select the connector). The required environment variables for this connector are:
OTTER_API_KEY
Step 1 — Get credentials
Enterprise workspaces only: Otter offers its Public API only to Enterprise workspaces. On other Otter plans the Developer tab is missing and this connector cannot connect. If your Enterprise workspace does not show the tab, your Otter account manager can turn it on.
- Sign in at otter.ai, click Integrations in the left navigation and open the Developer tab.
- Click Create key and copy the key at once: Otter shows it only once. Each user can have two keys.
- Paste the key into
OTTER_API_KEYas Otter shows it, without Bearer in front. - Check the setup with
otter_list_conversations.
Who sees what: the key acts as its user. The connector sees the conversations that user owns, and those shared with them when the AI asks for shared conversations or reads a channel. Share the MCP server only with people who may see those meetings.
Recordings and transcripts are personal data: they hold what colleagues, customers and other participants said, with their names and email addresses. The AI fetches only what your question needs, prefers the summary and action items over the full transcript, and does not copy transcripts or audio links to other systems unless you ask for it.
How to work: otter_list_conversations lists conversations newest first, optionally with those shared with you or only one channel from otter_list_channels, and pages on with a cursor. The API has no search and no date filter, so the AI pages back until it reaches the dates you asked about. otter_get_conversation reads Otter's summary, the action items with assignee and status, the insights by topic and the outline. otter_get_transcript returns what was said, 300 speaker blocks at a time; for longer meetings otter_get_transcript_more adds blocks 301 to 600, and beyond that the AI relies on the summary and outline. otter_list_channel_members and otter_get_workspace show people and the workspace.
Audio links: otter_get_audio_link returns a download link for the MP3 recording. The link opens the recording itself, so the AI gives it only to you.
Import asks first: otter_import_recording creates a new conversation from a recording at a public link (for example a call from your dialer), so Otter transcribes and summarises it. The AI confirms the link and the name with you and waits for your yes. Nothing is switched off at install, and no tool edits, shares or deletes conversations.
Limits and common errors: Otter allows 10 requests a second; more answer 429, so wait a moment. 401 unauthorized: the key is wrong or deleted, or the workspace has no Public API; create a new key under Integrations → Developer. 403: the key was accepted but the request refused; check that the workspace has the Public API turned on. 404: wrong conversation or channel id, or you cannot see it; on otter_get_workspace it means you are not in a workspace. 400: a parameter is invalid (a limit outside 1 to 100, or a cursor from another list).
The connector does not search by keyword or date (the API has neither), share or edit conversations, or set up webhooks (those are configured in Otter's workspace settings).
Step 2 — Install the adapter
curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d
Step 3 — Add the connector in ChatGPT
Your server URL: in AnythingMCP, open MCP Servers → the server this connector is on, and copy its URL (
https://cloud.anythingmcp.com/mcp/…). Use it wherever this guide showsYOUR_SERVER_ID.
ChatGPT adds MCP servers as plugins on its Plugins page, on the web and on paid plans. Developer mode is no longer needed. In a Business workspace only admins and owners can add one; in Enterprise and Edu an admin can grant the Create plugins with MCPs permission.
- Go to chatgpt.com/plugins, click +, then Add custom MCP server.
- Enter a name, e.g. Name:
Otter.ai. - Under Connection, enter the Server URL
https://cloud.anythingmcp.com/mcp/YOUR_SERVER_IDand keep Authentication on OAuth. - Select I understand and want to continue, then Create as a plugin. When ChatGPT asks, sign in to AnythingMCP and allow access.
- Install the new plugin from your personal plugins, then type @ in a chat and pick it.
Available tools
| Tool | What it does |
|---|---|
otter_get_workspace | Read the Otter workspace of the API key's user: id, name, owner, number of members, handle and type |
otter_list_channels | List the Otter channels of the user, alphabetically: id, name, number of members, owner and visibility (private or workspace) |
otter_list_channel_members | List the members of one Otter channel, alphabetically: user id, name and email |
otter_list_conversations | List Otter conversations (recorded meetings), newest first: id, title, date, link, owner, calendar guests, join link and Otter's summary |
otter_get_conversation | Read one Otter conversation without its full text: title, date, owner, guests, who it is shared with, Otter's summary, the action items… |
otter_get_transcript | Read what was said in an Otter conversation, speaker blocks 1 to 300: each block is the speaker name, the time into the recording and what… |
otter_get_transcript_more | Read speaker blocks 301 to 600 of a long Otter conversation, after otter_get_transcript showed totalBlocks over 300 |
otter_get_audio_link | Get a download link for the MP3 recording of an Otter conversation |
otter_import_recording | Create a new Otter conversation from a recording at a public link (for example a call from a dialer), so Otter transcribes and summarises it |
Example prompts
- "Summarise my last three Otter conversations and list the open action items with their owners."
- "What insights and action items came out of Tuesday's product review in Otter?"
- "Which conversations were shared in the Sales channel in Otter this week, and who is in that channel?"
- "Read the transcript of the interview with Julia Weber and pull out every quote about pricing."
- "Import the recording at https://example.com/calls/support-call.mp3 into Otter as Support call 14 October. Confirm the link and the name with me first."
FAQ
Does the free ChatGPT plan work with custom MCP servers? OpenAI documents custom MCP servers for paid plans only; the Free plan is not on its list.
Next steps
Was this guide helpful?
