Skip to content

meeting-agents MCP server

ai.mainroom/meeting-agentsmainroom.aistreamable-httpv1.0.0Scanned

AI agents that join live Google Meet, Teams, and Zoom calls as speaking participants.

8 tools: 1 do something that can't be undone, 3 change things and 4 only read.

Tools: 0 delete, 1 irreversible, 3 write, 4 read.
  • Reads private data · 2
  • Reads text others write · 0
  • Can send data out · 1
  • No findings

Tools listed without sign-in. Remote posture

With Anlyon

1 tool here does something that can't be undone. Govern it.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the meeting-agents MCP server can do

The meeting-agents MCP server is published by mainroom.ai, a domain the MCP Registry verified. It has 8 tools. It can do things that cannot be undone, such as sending, paying or publishing (send_demo_agent) and create or change data (3 tools: about_mainroom, join_meeting_as and delegate_to_meeting), but cannot delete data. 4 tools only read.

2 tools read private data (meeting_status and list_relationships); 1 tool can send data out of your system (send_demo_agent).

None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 11, 2026, version 1.0.0.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

8 tools

Can't be undone 1Claude Code: ask
  • send_demo_agentSend Mainroom's open demo agent into a live meeting RIGHT NOW.Can send data outNo findings

    Why it is in this group

    • The name "send_demo_agent" says "send": a word for sending, and what is sent cannot be taken back.

    What it can reach

    • outboundThe name "send_demo_agent" says "send": it sends or publishes something to other people.
    Description, as the server wrote it
    Send Mainroom's open demo agent into a live meeting RIGHT NOW. Give it a Google Meet, Microsoft Teams, Zoom, or Webex meeting URL; the agent appears in the meeting lobby within about a minute and a human in the meeting must admit it. It introduces itself, answers questions by voice, presents slides, and can run a live browser demo. Use when your user wants an AI participant in a meeting that is happening now. No account needed.
Changes things 3Claude Code: ask
  • about_mainroomWhat Mainroom is: AI agents that join live Google Meet / Teams / Zoom calls as real participants (voice, slides, live demos, research, memory, follow-up emails).No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    What Mainroom is: AI agents that join live Google Meet / Teams / Zoom calls as real participants (voice, slides, live demos, research, memory, follow-up emails). Explains how humans invite agents, and how AI assistants can dispatch the demo agent, join a meeting themselves, or send the user's delegate and follow the call.
  • delegate_to_meetingDELEGATE: send the user's own stand-in into a live meeting they can't attend.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    DELEGATE: send the user's own stand-in into a live meeting they can't attend. The delegate joins as a named participant, introduces itself as standing in for them, listens, relays their positions from the brief, asks the questions they need answered at natural moments, and hands every decision back to them. Afterwards a report of what they missed lands in their inbox. Needs the user's Mainroom API key (History page at https://app.mainroom.ai). Returns a session id: poll meeting_status with it to follow the call live (transcript, running summary, answers so far) and to get the final summary once it ends. Use when your user says 'cover this meeting for me', 'I can't make the 3pm', 'send someone in my place'.
  • join_meeting_asEMBODIMENT: join a live meeting AS YOURSELF (or as a persona you define).No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    EMBODIMENT: join a live meeting AS YOURSELF (or as a persona you define). This gives an AI agent a meeting body — a named participant with a camera tile and a voice that joins the call, speaks when addressed, presents slides, shows images, and reads/posts meeting chat. You supply the name and the persona instructions (who you are, what you're there to do, what you know); Mainroom supplies the body. The participant appears in the meeting lobby within ~a minute and a human must admit it. Ephemeral: no account, no memory, nothing emailed afterwards. Use when your user says 'join my call', 'be in the meeting', or you need to talk to people in a live meeting to complete a task.
Only reads 4Show 4 read-only tools
  • get_relationship_historyHISTORY: everything the user's Mainroom agents recorded with one account, team, or person: what is still open (commitments with owners and due dates, unanswered questions, risks), …No findings

    Why it is in this group

    • The name "get_relationship_history" says "get": a word for reading.
    Description, as the server wrote it
    HISTORY: everything the user's Mainroom agents recorded with one account, team, or person: what is still open (commitments with owners and due dates, unanswered questions, risks), the proposed / approved follow-up actions, and a session-by-session timeline with summaries, decisions, and pre-call briefs. Use it to answer 'where are we with Acme', 'what did we promise them', 'what did the team decide last week', or to prep for the next call. Needs the user's Mainroom API key.
  • list_agent_templatesList Mainroom's built-in agent types (facilitator, timekeeper, sales support, mock interviewer, ...).No findings

    Why it is in this group

    • The name "list_agent_templates" says "list": a word for reading.
    Description, as the server wrote it
    List Mainroom's built-in agent types (facilitator, timekeeper, sales support, mock interviewer, ...).
  • list_relationshipsHISTORY: list the customer accounts, recurring team meetings, and people that the user's own Mainroom agents have sat in meetings with — with session counts, when they last met, an…Reads private dataNo findings

    Why it is in this group

    • The name "list_relationships" says "list": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "meeting": it reads calendars.
    Description, as the server wrote it
    HISTORY: list the customer accounts, recurring team meetings, and people that the user's own Mainroom agents have sat in meetings with — with session counts, when they last met, and how many commitments / open questions / risks are still open. Needs the user's Mainroom API key (shown on the History page at https://app.mainroom.ai — they paste it once). Follow with get_relationship_history for the detail.
  • meeting_statusFOLLOW A MEETING: what the user's agent (a delegate, or any of their agents) is seeing in a meeting right now, or what it recorded once the call ended — the last lines of transcrip…Reads private dataNo findings

    Why it is in this group

    • The name "meeting_status" says "status": a word for reading.

    What it can reach

    • private-dataThe name "meeting_status" says "status" and "meeting": it reads calendars.
    • private-dataThe description's first sentence says "meeting": it reads calendars.
    Description, as the server wrote it
    FOLLOW A MEETING: what the user's agent (a delegate, or any of their agents) is seeing in a meeting right now, or what it recorded once the call ended — the last lines of transcript, the running summary, answers to the user's questions so far, decisions and commitments, and whether the bot is still waiting in the lobby. Call without a session id to list the user's recent and live sessions. Needs the user's Mainroom API key.

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged untrusted-input.

Reads private data2
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write0
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out1
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 18:26 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
4
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • HighThe whole server

    The endpoint served tools/list without credentials, and the list has 4 tools that change things (1 irreversible, 3 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://mcp.mainroom.ai/mcp: tools/list returned 8 tools to an anonymous client, including send_demo_agent (irreversible), about_mainroom (write), join_meeting_as (write), delegate_to_meeting (write)

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

No check matched anything in these tool definitions. That is not a safety guarantee.

History

Changes to its tool definitions and to its remote endpoint, newest first.

No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.

Install

Pinned to version 1.0.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http meeting-agents https://mcp.mainroom.ai/mcp
claude mcp add --transport http meeting-agents https://mcp.mainroom.ai/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__meeting-agents__list_agent_templates",
      "mcp__meeting-agents__meeting_status",
      "mcp__meeting-agents__list_relationships",
      "mcp__meeting-agents__get_relationship_history"
    ],
    "ask": [
      "mcp__meeting-agents__about_mainroom",
      "mcp__meeting-agents__send_demo_agent",
      "mcp__meeting-agents__join_meeting_as",
      "mcp__meeting-agents__delegate_to_meeting"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as meeting-agents. Review it before you use it.

Capability tags behind 3 rules
  • mcp__meeting-agents__send_demo_agent outbound
  • mcp__meeting-agents__meeting_status private-data
  • mcp__meeting-agents__list_relationships private-data

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/ai.mainroom/meeting-agents/verify?hash=d5284a27ebe56afdd02f49f3570a13f0e8cd09f306b67fc1ca3e448117ee46e9" | jq -e '.changed == false'

Pinned hash d5284a27ebe5, from Nov 11, 2026, 20:17 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/ai.mainroom/meeting-agents/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for ai.mainroom/meeting-agents
[![Anlyon Radar](https://radar.anlyon.com/badge/ai.mainroom/meeting-agents.svg)](https://radar.anlyon.com/servers/ai.mainroom/meeting-agents)

Source: MCP Registry. Synced Oct 21, 2026, 17:42 UTC.

meeting-agents MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the meeting-agents MCP server safe?

No scan can promise that. What its definitions show: 8 tools, 4 of which can change, send or delete something, and 0 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the meeting-agents MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the meeting-agents MCP server do things that cannot be undone?

Yes. 1 tool can send, pay, publish or merge: send_demo_agent.

What tools does the meeting-agents MCP server have?

8 tools: 1 irreversible, 3 write, 4 read. send_demo_agent, about_mainroom, join_meeting_as, delegate_to_meeting, list_agent_templates, meeting_status, list_relationships and get_relationship_history.

Does the meeting-agents MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the meeting-agents MCP server send data to outside services?

1 tool can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): send_demo_agent.

Has the meeting-agents MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 11, 2026.