Tenali MCP Server

Connect Tenali to Claude, ChatGPT, Perplexity, Gemini and Cursor with the Tenali MCP server — a read-only Model Context Protocol connector for your meetings, transcripts, action items and Question Intelligence.

The Tenali MCP server lets an AI assistant read your Tenali data directly. Instead of pasting transcripts into a chat window, you connect Tenali once and ask questions in plain language — the assistant looks everything up itself.

It's read-only, scoped to permissions you approve at sign-in, and included with any Tenali plan that has AI chat — there's no separate MCP add-on.

What MCP Is

MCP (Model Context Protocol) is an open standard for connecting an AI assistant to another product's data. Tenali runs an MCP server; Claude, ChatGPT, Perplexity, Gemini and Cursor are MCP clients that know how to talk to it. Connect the two and the assistant gains a set of tools it can call on your behalf.

Your Server URL

Every connection uses the same endpoint:

https://api.tenali.ai/api/mcp

You'll also find it, with a copy button, in the web app under Settings → AI Assistants.

What You Can Ask

Once connected, your assistant can answer questions grounded in your own meetings and knowledge base:

Customer questions — "What objections came up most last quarter?" pulls your Question Intelligence, including the answer Tenali captured for each one.
Meeting search — "Summarize every call we've had with Acme" searches summaries, briefs, key points and next steps.
Action items — "What did I commit to on Tuesday?" lists follow-ups with the call each came from.
Account review — "Which accounts have gone quiet?" returns momentum derived from your meetings.
Knowledge base — "What does our security doc say about SOC 2?" searches your docs, FAQs and training sources.
Transcripts — "Pull the exact quote where pricing came up" reads raw transcript text, when you grant that permission.

See the Tool Reference for the complete list of what an assistant can call.

What You Need

A Tenali account on a plan that includes AI chat. There's no separate MCP subscription or add-on — access mirrors your existing plan.
A plan on the assistant's side that supports custom connectors or apps. Claude Free includes one connector. In ChatGPT, full MCP support is in beta on Business, Enterprise and Edu, while Pro can connect MCP servers with read/fetch permissions — enough for Tenali, since every Tenali tool is read-only. Gemini's custom apps have their own eligibility rules; see Connect Perplexity, Gemini, Cursor & Other Clients.

Connect Your Assistant

1
Copy your server URL
In Tenali, open Settings → AI Assistants and click Copy next to the server URL.
2
Add it to your assistant
Follow the guide for your client — Claude, ChatGPT, or Perplexity, Gemini, Cursor and others.
3
Approve what it can access
Sign in with your Tenali account. On the approval screen, untick any of the optional permissions you'd rather keep out — Meeting summaries and Question Intelligence are required and stay selected — then click Allow access.
4
Confirm it's connected
Back in Tenali, Settings → AI Assistants lists the connection under Your connections. Ask your assistant something like "list my recent Tenali meetings" to verify it's working.

Security in Brief

Every tool is read-only — nothing the assistant calls can edit, delete or send anything. Your identity is pinned when you approve the connection and re-checked on every request, so an assistant can never read more than you can see in Tenali. Raw transcripts and organization-wide access are separate opt-ins.

Full detail is in Permissions & Security.