# Mainroom — full site text Generated 2026-09-17. Mainroom builds AI agents that join live meetings on Google Meet, Microsoft Teams and Zoom as participants. See https://mainroom.ai/llms.txt for the summary. --- # What is an AI meeting agent? URL: https://mainroom.ai/blog/what-is-an-ai-meeting-agent/ Updated: 2026-09-17 An AI meeting agent is an AI system that joins a live meeting as a participant and takes part in it: it listens to the room, speaks when addressed, puts material on the shared screen, looks things up, and acts on what is decided. That is the difference from an AI notetaker, which records a meeting and produces a summary afterwards. The short definition An AI meeting agent has four properties that a notetaker does not: 1. It is in the room. It has a name in the participant list, a camera tile, and a voice. People address it the way they address a colleague. 2. It acts during the meeting, not after. It answers a question, pulls up a slide, runs a demo, or opens a poll while the conversation is happening. 3. It has manners. It waits for a gap, stops when a person talks over it, goes on mute when told to, and raises a hand instead of interrupting. 4. It has tools. It can search the web and the team's own systems, present on the shared screen, and draft follow-ups that a human approves. The recap at the end is a by-product. Notetakers make notes; agents make the meeting itself go differently. What an AI meeting agent can do in a call Here is what a Mainroom agent does today, in a real Google Meet, Microsoft Teams or Zoom call, with nothing installed on the platform: - Speak and listen. Joins as a participant with a camera tile and a voice. Answers when addressed, waits for a gap, and never talks over a person. - Slides and pages on the shared screen. Puts up text slides, comparison tables, charts, timelines and small calculators on the meeting's shared screen, and takes them down when asked. - Whiteboard. Teaches on a shared whiteboard it draws on step by step: boxes, arrows, notes, highlights. Questions about the drawing are answered on the drawing. - Live browser demos. Opens a real browser on the shared screen, follows instructions from the room ("click the pricing tab", "scroll down"), and narrates. - Lookups in connected tools and the web. Runs a background search in the team's connected tools (any MCP server) or the web, and reports when there is a gap in conversation. - Generated images. Generates an image from a description and puts it on the shared screen. - Anonymous intake. Opens an anonymous response link for the room, shows a live count, and synthesizes the responses on request. - Meeting chat and private notes. Posts links and recaps to the meeting chat; sends a private note to one person on request. - Meeting ledger. Logs decisions, commitments with owners and dates, open questions and risks the moment they are confirmed, and carries them to the next meeting with the same people. - Proposed actions. Drafts CRM notes, tickets and follow-ups as proposals the owner approves before anything is written to a system of record. - Live coaching. Sends private nudges and answers private questions on a companion page only the inviter can see. - Memory. Remembers people and prior sessions across calls, per agent, as a switch the owner controls. - Post-call email. Emails a role-appropriate recap after the call: a coaching debrief, an account summary, or a team recap. - Floor control by voice. Goes on mute when told to and stays silent until asked to come off mute. Raises a hand instead of interrupting. How an AI meeting agent joins a meeting On Mainroom the agent has its own email address. You add that address to a calendar invite, or forward the invite to it, and the agent joins at the start time. It appears in the lobby and the host admits it, exactly like a guest. Captions on the platform tell it who is speaking by name. Under the hood the agent runs on a voice model that listens and speaks in real time (OpenAI's realtime voice models, in our case), with a gate in front of the model that decides when a reply is wanted at all. That gate is most of the work. A model that answers everything it hears is unusable in a room of people talking to each other, so the agent tracks who was addressed, who is in the meeting, whether it is a one-on-one or a room, and whether it has been told to be quiet. When you would want one The honest list, from our own use: - Customer calls, where an agent remembers the account, logs commitments both ways, and coaches the account owner privately without the customer seeing anything. - Recurring team meetings, where the agent confirms decisions, asks for owners and dates as actions land, and brings back last week's open items. - Teaching and explaining, where the agent draws on a whiteboard step by step instead of describing a diagram in the air. - Demos, where the agent drives a real browser on the shared screen and follows instructions from the room. - Getting the quiet half of the room heard, with an anonymous intake link and a synthesis of the responses. When you would not An agent is a participant, and participants have a cost: attention. If a meeting is a private conversation between two people who know each other well, a notetaker that stays out of the way may be the better tool. Mainroom's Silent Coach template exists for exactly that case: it joins as a plain notetaker, never speaks, and coaches the person who invited it privately. What to look for when evaluating one - Does it join the platforms you use, as a participant, without a plugin? - Can you tell it to be quiet, and does it stay quiet? - Does it act on the shared screen, or only in a side panel? - What can it connect to, and does anything get written to your systems without approval? - Who sees the private coaching, if there is any? - Can you build your own agent, with its own persona, tools and manners, or only use a fixed assistant? Mainroom is one answer to these questions. The comparison pages go through the alternatives. Q: Is an AI meeting agent the same as an AI notetaker? A: No. A notetaker records the meeting and produces notes afterwards. A meeting agent takes part in the meeting while it is happening: it answers, presents, looks things up, and gathers input. Most agents also produce notes, but that is not what makes them agents. Q: Does an AI meeting agent need a plugin or app installed? A: Not on Mainroom. The agent joins Google Meet, Microsoft Teams or Zoom as a regular participant from a calendar invite, and the host admits it from the lobby. Q: Can everyone in the meeting see the agent? A: Yes. It appears in the participant list with its own name and camera tile, and the platform shows its standard notice. It never joins silently. --- # AI notetaker vs. AI meeting agent: what is the difference? URL: https://mainroom.ai/blog/ai-notetaker-vs-ai-meeting-agent/ Updated: 2026-09-17 The two are often confused because both show up in the participant list with a bot name. The difference is what happens during the meeting. An AI notetaker listens and writes; an AI meeting agent listens, speaks, and acts. Side by side AI notetaker AI meeting agent --------- Joins the call as a participant Yes Yes Produces a transcript and summary Yes Yes Answers a question during the call No Yes, when addressed Puts a slide, chart or table on the shared screen No Yes Draws on a whiteboard to explain something No Yes Runs a live browser demo the room can steer No Yes Looks something up in your tools mid-call No Yes, in the background Opens an anonymous poll for the room No Yes Logs decisions and commitments as they are confirmed After the call During the call, and carries them forward Drafts CRM notes and follow-ups for approval Sometimes, after the call Yes, as proposed actions Private coaching for one person during the call No Yes, on a companion page Can be told to be quiet Not applicable Yes, by voice Remembers people across meetings Sometimes Yes, per agent What a notetaker is for A notetaker's job is fidelity. Record everything, attribute it correctly, summarize it well, make it searchable later. The best ones are very good at this, and the category is mature. If what you need is a reliable record, a notetaker is the right tool and a cheaper one. What an agent is for An agent's job is the meeting's outcome. The three places it earns its seat: 1. Decisions stop waiting. The number, the date, the document, the answer from your tools arrives in the call rather than in a follow-up three days later. 2. Meetings end actionable. Owners and dates are asked for as actions land, a recap is in the chat before people leave, and follow-ups are queued for approval instead of lost in someone's notes. 3. Everyone gets heard. An anonymous intake link for the quiet half of the room, a nudge when one person has had the last five minutes, and an agent that waits its turn. The trade-off An agent asks for attention that a notetaker does not. Our own rule, written into the default meeting manners: speak only when addressed, wait for a gap, stop when a person talks over you, and go on mute when told to. On customer calls the agent is silent unless its owner names it. A notetaker never has to be told any of this. Which one to pick - You want a record. Notetaker. - You want the meeting to go differently. Agent. - You want private coaching without a visible AI in the room. A silent agent: Mainroom's Silent Coach joins as "Mainroom Notetaker", never speaks, and coaches the inviter privately. The Mainroom comparisons look at specific products in each category. Q: Do I need both a notetaker and a meeting agent? A: Usually not. A meeting agent produces a recap and a ledger of decisions and actions as well, so it covers what a notetaker does. Teams that already have a notetaker they like can run a Mainroom agent alongside it; the agent does not need to be the recorder of record. Q: Which is better for a one-on-one? A: For a private one-on-one a silent agent or a notetaker is usually right. Mainroom's Silent Coach template joins as a notetaker, never speaks, and coaches the person who invited it privately. --- # How to invite an AI agent to a Google Meet, Microsoft Teams or Zoom call URL: https://mainroom.ai/blog/how-to-invite-an-ai-agent-to-google-meet-teams-or-zoom/ Updated: 2026-09-17 An AI agent joins a meeting on Mainroom the same way a person does: from the calendar invite. There is no plugin, no bot marketplace and nothing to install on Google Meet, Microsoft Teams or Zoom. Here is the whole process. Before you start You need a Mainroom account (free during beta) and a meeting on Google Meet, Microsoft Teams or Zoom. The agent works best when the platform's captions are on, because captions are how it knows who is speaking by name. Step 1: build the agent In the Studio at app.mainroom.ai, start from a template or describe the agent in plain language and let the builder draft it. Each agent has: - A persona: a name, a one-line role, and how it should behave. - A voice, from a set of natural voices. - Tools: slides, whiteboard, live browser demos, images, meeting chat, vision of shared screens, and connections to your tools through MCP servers. - Meeting manners, written in plain language, plus a default floor state: open, hand-raise, or muted. - Switches for memory, the post-call recap, live coaching, an AI-disclosure notice, and discreet mode. Step 2: give it an address and an invite policy Every agent has its own email address at mainroom.ai. Decide who can invite it: only you, or anyone in your company's email domain. Colleagues do not need a Mainroom account to invite an agent you built. Step 3: add it to the calendar invite Add the agent's address as a guest on the calendar event, or forward the invite to it. That is the entire integration. Step 4: admit it from the lobby At the start time the agent joins and appears in the lobby. The host admits it like a guest. It shows up in the participant list with its name and a camera tile, and the platform displays its standard recording notice, so everyone in the room can see it is there. Step 5: talk to it Address the agent by name. Some things people ask in the first five minutes: - "Steve, put up a slide with the three options." - "Can you teach me how OAuth works?" (it draws on the whiteboard) - "Show us the pricing page." (it opens a live browser demo) - "Look up the last five tickets from this customer." - "Let's get everyone's take on this anonymously." - "Go on mute for a bit." and later "Steve, come off mute." - "You can drop off now." In a one-on-one call it assumes you are talking to it. In a room it answers when addressed and otherwise listens. Step 6: use the companion page Whoever invited the agent gets a private link to my.mainroom.ai. During the call it shows live indicators (talk share, questions asked, longest stretch, filler words), short private nudges, and a private chat with the agent. Nothing on it appears in the meeting. Step 7: after the call The agent emails a recap matched to its role: a coaching debrief, an account summary, or a team recap with the ledger of decisions, commitments, open questions and risks. Anything it wants to write to a CRM, a ticket or a follow-up email arrives as a proposed action you approve or discard. Troubleshooting - The agent joined but does not know who is speaking. Turn on captions on the platform. - It cannot share its screen. Allow participants to share their screen in the meeting settings. - It answered something meant for the room. Tell it to go on mute, or tighten its manners in the Studio. Customer-call templates default to silent unless named. - You want it gone. Say so, or have the host remove it. --- # Meeting manners for AI agents: what we learned putting one in real calls URL: https://mainroom.ai/blog/meeting-manners-for-ai-agents/ Updated: 2026-09-17 The hardest part of building an AI participant is not making it speak. It is making it stay quiet. These are first-hand notes from putting Mainroom agents into real Google Meet and Teams calls over the past months, with the rules that survived. Rule 1: being mentioned is not being addressed "I invited Steve to this meeting, and Steve is an AI agent" is not a question for Steve. Neither is "so Steve is supposed to know when not to jump in?" The agent has to tell talking to it from talking about it, and the default has to be silence. Our gate treats a name in the third person as a mention and stays quiet; a question with the name in the second person is an address. Rule 2: a one-on-one is a different meeting When there is one person on the call with the agent, everything they say is probably for it, including thinking aloud. "Help me understand." followed by a pause is the start of a question, not a question. The agent waits a few seconds longer for the rest before answering. In a room, the same words are for the room. We know how many people are present from the platform's participant list, and the agent's whole posture changes with that number. Rule 3: mute means mute "Go on mute" is an instruction, and the agent stays muted until someone says "come off mute". Its name alone does not wake it. "Steve, quick one even though you're muted" gets silence. That felt harsh in testing and turned out to be right: people mute the agent to run part of a meeting without it, and any exception becomes a way for it to creep back in. The take-down that often comes with a mute ("take that slide down and go on mute") is handled in the same breath. Rule 4: stop the instant a person talks over you The agent freezes its audio within about a hundred milliseconds of a person starting to speak over it. If the burst turns out to be a backchannel ("mm-hm", "right"), it resumes; if the person keeps going, it stops for good and remembers where it was, so "sorry, go on" continues rather than restarting. Two failure modes we fixed along the way: an interruption that started inside the short echo guard used to be ignored for the whole burst, and a long interruption that the transcriber shortened to "Hey Steve." used to be treated as a backchannel. Both are now handled from the audio, not the words. Rule 5: a bare hail gets a bare reply "Hey Steve." wants "Yeah?", not a status report about being ready to help. The reply should be under eight words and then wait. Rule 6: findings wait for a gap A background lookup comes back in ten to thirty seconds. The agent does not announce it mid-sentence. It holds the result until there is a pause, then says it, and if asked to put it on screen it does that rather than reading it aloud. Rule 7: demos narrate themselves, and the room steers When a live browser demo is on the shared screen, "click the pricing tab" and "scroll down" are for the agent even without its name. The narration sentence for each action starts before the click lands, so the room sees the action mid-sentence rather than after the words. If a person starts talking, the next narration line waits. Rule 8: customer calls are silent unless the owner names you On a customer call the agent supports the account owner. In front of the customer it stays silent unless the owner brings it in, never volunteers internal information, and does its real work privately: noticing commitments, unanswered questions and risk signals, and telling the owner quietly on the companion page. Rule 9: never claim an action you did not take "Closing it now" with nothing closed is worse than silence. Actions the agent has tools for (mute, leave, take down the screen, close a demo) are executed deterministically by the application when the words are unambiguous, and the model is told to say only what it delegated. A small audit after each turn checks that every part of a two-part request was actually done. Rule 10: write the manners in plain language, per agent Every Mainroom agent has a manners field its owner writes in plain English. "Silent unless your owner names you." "Speak only for process: a decision to confirm, an action needing an owner." "Engage actively, this is practice." The defaults above are the floor; the owner sets the ceiling. What we measure We run every change through a simulator that plays scripted meetings through the real agent page with synthetic voices, and score it: did it answer when addressed, stay silent when mentioned, mute when told to, resume after a backchannel. On the realtime engine the median time from the end of a person's sentence to the agent's first audio is about 2.5 seconds; on the newer full-duplex engine it is about 1.3 seconds. Those numbers are from our own runs in September 2026 and move with every model release. --- # An AI that teaches on a whiteboard, inside the meeting URL: https://mainroom.ai/blog/ai-whiteboard-teaching-in-meetings/ Updated: 2026-09-17 Ask a Mainroom agent "how does OAuth work?" and it does what a good colleague does at a whiteboard: it draws two boxes, says a sentence, draws an arrow, says another, and keeps going until the picture is complete, then asks whether it makes sense. What it looks like The board appears on the meeting's shared screen. Elements arrive as marker strokes: boxes and circles draw themselves, labels write themselves in, arrows shoot from one thing to another, highlights sweep across a word, and anything erased gets scribbled out before its replacement is written. The camera frames what is on the board, zooming in while it is sparse and out as it fills. The agent draws one or two elements per step, explains them in two or three sentences with an example, and continues without being asked. Questions about the drawing are answered on the drawing: "which of those is the one attackers go after?" gets a red ring around the box, not a paragraph. Why a whiteboard and not a slide A slide is a finished thing. It arrives all at once, and the room reads ahead of the speaker. A drawing grows at the pace of the explanation, so the picture and the words stay together. That is why people teach at whiteboards, and it is why an agent that only had slides felt like it was reading a document at you. When the agent reaches for it The agent treats "how does X work", "walk me through", "what's the difference between", "help me understand" and similar as teaching asks with structure to them, and answers on the board. It does not use the board for a plain list or a recap (that is a slide), for a one-line factual answer, or on a customer call unless its owner asks. How it stays readable Some rules the board enforces on the agent's behalf: - Labels are one to five words, notes under ten. - New elements flow left to right and top to bottom unless the agent places them, and never land on an arrow. - A new note never writes over an existing one: the old note is erased first. - Arrows run behind boxes and bow around a box they would otherwise cross. - About ten elements is a full board. Past that the agent clears and starts a fresh titled board, or builds in the empty half. What it is built on The drawing is a small vector model the agent addresses with plain operations (box, text, arrow, circle, highlight, check, erase, clear), rendered as animated SVG on a page the meeting bot shares. The agent's reasoning model decides what to draw; the page decides where it fits and how it animates. That split is what keeps the model's attention on the lesson. Try it: invite the demo agent at demo@mainroom.ai to a call and ask it to teach you something. Q: Can I ask the agent to draw something specific? A: Yes. "Draw the flow", "circle the risky step", "add the token exchange", "clear the board and start over" all work. The agent also decides on its own when a question would be better answered with a drawing than with words. Q: Does the whiteboard work on Google Meet, Teams and Zoom? A: Yes. The board is shared through the platform's screen share, so it works wherever the agent can share its screen. --- # Blog URL: https://mainroom.ai/blog/ Updated: 2026-09-17 - What is an AI meeting agent? — An AI meeting agent is an AI participant in a live meeting: it listens, speaks, presents, and acts, unlike a notetaker that records and summarizes. Definition, capabilities, and how one joins a call. - AI notetaker vs. AI meeting agent: what is the difference? — AI notetakers record and summarize. AI meeting agents participate: they answer, present, demo, gather input, and act. A side-by-side comparison with a table. - How to invite an AI agent to a Google Meet, Microsoft Teams or Zoom call — Step by step: build an agent, give it an email address, add it to a calendar invite, admit it from the lobby, and talk to it. No plugin or install on the meeting platform. - Meeting manners for AI agents: what we learned putting one in real calls — The hardest part of an AI meeting participant is knowing when not to speak. First-hand notes on addressing, mentions, one-on-one versus rooms, mute, interruptions and demos. - An AI that teaches on a whiteboard, inside the meeting — How a Mainroom agent explains a concept by drawing it: one or two elements at a time, talking over the marker, answering questions on the drawing. What it looks like and why it beats a slide. --- # Mainroom vs. Otter.ai URL: https://mainroom.ai/compare/mainroom-vs-otter-ai/ Updated: 2026-09-17 In one sentence: Otter.ai transcribes and summarizes meetings. Mainroom puts an AI participant in the meeting that speaks, presents, draws and acts. What Otter.ai does - Live transcription during the meeting, with speaker identification. - Automated meeting notes and summaries after the call, and a chat over your notes. - A bot (OtterPilot) that joins Zoom, Google Meet and Microsoft Teams to record. - Search across all past meetings. What Mainroom does Mainroom builds AI agents that join Google Meet, Microsoft Teams and Zoom as participants from a calendar invite. In the call an agent speaks when addressed, puts slides and tables on the shared screen, teaches on a whiteboard, runs live browser demos the room can steer, looks things up in your tools, opens anonymous intake for the room, logs decisions and commitments as they are confirmed, and coaches the person who invited it privately. After the call it emails a recap and queues proposed actions for approval. Side by side Otter.ai Mainroom --------- Category AI notetaker AI meeting agent Joins Meet, Teams and Zoom as a participant Yes Yes Transcript and summary Yes, core product Yes, as a by-product (recap email, ledger) Speaks in the meeting when addressed No Yes Puts slides, tables and charts on the shared screen No Yes Whiteboard teaching No Yes Live browser demo steered by the room No Yes Lookups in your tools mid-call No Yes, any MCP server plus web search Anonymous intake for the room No Yes Proposed actions to CRM and tickets with approval No Yes Private live coaching for the inviter No Yes Build your own agent with persona, tools and manners No Yes Pricing Free tier and paid plans (see otter.ai) Free during beta Which to choose Pick Otter.ai if what you need is a dependable transcript and searchable notes across every meeting. Pick Mainroom if you want the meeting itself to go differently: answers in the call, material on the screen, input from everyone, and follow-ups queued for approval. Many teams run both; a Mainroom agent does not need to be the recorder of record. A note on accuracy Details about other products on this page come from their public product pages as of September 17, 2026 and may have changed; check the vendor's site for current features and pricing. Mainroom details are current as of the same date. Corrections: hello@mainroom.ai. Q: Can I use Otter.ai and Mainroom together? A: Yes. A Mainroom agent is one more participant in the meeting and does not need to be the recorder of record. Teams that like their current notetaker keep it. Q: Does Mainroom produce a transcript and summary too? A: Yes. Every call ends with a recap email matched to the agent's role and a ledger of decisions, commitments, open questions and risks that carries to the next meeting with the same people. Q: Does anything get written to my CRM without approval? A: No. Writes to systems of record arrive as proposed actions that the agent's owner approves or discards. --- # Mainroom vs. Fireflies.ai URL: https://mainroom.ai/compare/mainroom-vs-fireflies-ai/ Updated: 2026-09-17 In one sentence: Fireflies.ai is a meeting recorder and note-taker with integrations. Mainroom is an AI participant that speaks, presents and acts during the call. What Fireflies.ai does - A bot (Fred) that joins meetings on the major platforms to record and transcribe. - Summaries, action items and topic tracking after the call. - Integrations that send notes and action items to CRMs, project tools and Slack. - Search and a chat assistant over the meeting archive. What Mainroom does Mainroom builds AI agents that join Google Meet, Microsoft Teams and Zoom as participants from a calendar invite. In the call an agent speaks when addressed, puts slides and tables on the shared screen, teaches on a whiteboard, runs live browser demos the room can steer, looks things up in your tools, opens anonymous intake for the room, logs decisions and commitments as they are confirmed, and coaches the person who invited it privately. After the call it emails a recap and queues proposed actions for approval. Side by side Fireflies.ai Mainroom --------- Category AI notetaker with integrations AI meeting agent Joins Meet, Teams and Zoom as a participant Yes Yes Transcript and summary Yes, core product Yes, as a by-product Sends notes to CRM and project tools Yes, automatically after the call As proposed actions the owner approves Speaks in the meeting when addressed No Yes Shared-screen slides, charts, whiteboard, demos No Yes Lookups in your tools mid-call No Yes, any MCP server plus web search Anonymous intake for the room No Yes Private live coaching for the inviter No Yes Remembers people and accounts across calls Archive search Yes, per agent, as a switch Build your own agent No Yes Pricing Free tier and paid plans (see fireflies.ai) Free during beta Which to choose Fireflies.ai is strong at getting a record out of every meeting and into the tools you already use. Mainroom is for teams that want an agent in the room: it answers, presents, and gathers input during the call, and its writes to your systems wait for a human. If your problem is "nobody reads the notes", an agent that asks for owners and dates while the meeting is still on is the more direct fix. A note on accuracy Details about other products on this page come from their public product pages as of September 17, 2026 and may have changed; check the vendor's site for current features and pricing. Mainroom details are current as of the same date. Corrections: hello@mainroom.ai. Q: Can I use Fireflies.ai and Mainroom together? A: Yes. A Mainroom agent is one more participant in the meeting and does not need to be the recorder of record. Teams that like their current notetaker keep it. Q: Does Mainroom produce a transcript and summary too? A: Yes. Every call ends with a recap email matched to the agent's role and a ledger of decisions, commitments, open questions and risks that carries to the next meeting with the same people. Q: Does anything get written to my CRM without approval? A: No. Writes to systems of record arrive as proposed actions that the agent's owner approves or discards. --- # Mainroom vs. Read AI URL: https://mainroom.ai/compare/mainroom-vs-read-ai/ Updated: 2026-09-17 In one sentence: Read AI adds engagement analytics to meeting notes. Mainroom puts a participant in the meeting and coaches one person privately while it happens. What Read AI does - Meeting notes and summaries from a bot that joins the major platforms. - Engagement, sentiment and talk-time analytics per meeting. - Coaching-style metrics reported after the call. - Search across meetings and other sources. What Mainroom does Mainroom builds AI agents that join Google Meet, Microsoft Teams and Zoom as participants from a calendar invite. In the call an agent speaks when addressed, puts slides and tables on the shared screen, teaches on a whiteboard, runs live browser demos the room can steer, looks things up in your tools, opens anonymous intake for the room, logs decisions and commitments as they are confirmed, and coaches the person who invited it privately. After the call it emails a recap and queues proposed actions for approval. Side by side Read AI Mainroom --------- Category AI notetaker with analytics AI meeting agent Joins Meet, Teams and Zoom as a participant Yes Yes Transcript and summary Yes Yes, as a by-product Talk-share and engagement indicators Yes, after the call Yes, live on the inviter's private companion page Private nudges during the call No Yes, only the inviter sees them Speaks in the meeting when addressed No Yes Shared-screen slides, whiteboard, demos No Yes Anonymous intake for the room No Yes Proposed actions with approval No Yes Silent mode (joins as a notetaker, never speaks, coaches privately) Notes only Yes, Silent Coach template Build your own agent No Yes Pricing Free tier and paid plans (see read.ai) Free during beta Which to choose If the value you want is a scorecard after each meeting, Read AI delivers that as a product. Mainroom delivers the same kind of indicators live, to one person, with nudges while there is still time to act on them, and it can also take part in the meeting. For managers who want coaching without a visible AI in the room, Mainroom's Silent Coach is the closest match. A note on accuracy Details about other products on this page come from their public product pages as of September 17, 2026 and may have changed; check the vendor's site for current features and pricing. Mainroom details are current as of the same date. Corrections: hello@mainroom.ai. Q: Can I use Read AI and Mainroom together? A: Yes. A Mainroom agent is one more participant in the meeting and does not need to be the recorder of record. Teams that like their current notetaker keep it. Q: Does Mainroom produce a transcript and summary too? A: Yes. Every call ends with a recap email matched to the agent's role and a ledger of decisions, commitments, open questions and risks that carries to the next meeting with the same people. Q: Does anything get written to my CRM without approval? A: No. Writes to systems of record arrive as proposed actions that the agent's owner approves or discards. --- # Mainroom vs. Zoom AI Companion, Microsoft Copilot in Teams, and Gemini in Google Meet URL: https://mainroom.ai/compare/mainroom-vs-zoom-ai-companion-copilot-and-gemini/ Updated: 2026-09-17 In one sentence: The native assistants live inside one platform and answer questions about the meeting. Mainroom is a cross-platform participant that speaks, presents and acts. What they do - Zoom AI Companion: meeting summaries, in-meeting questions about what was said, and other assistance inside Zoom. - Microsoft Copilot in Teams: recaps and in-meeting questions inside Teams, with a Microsoft 365 Copilot license. - Gemini in Google Meet: note-taking and summaries inside Meet for Google Workspace customers. What Mainroom does Mainroom builds AI agents that join Google Meet, Microsoft Teams and Zoom as participants from a calendar invite. In the call an agent speaks when addressed, puts slides and tables on the shared screen, teaches on a whiteboard, runs live browser demos the room can steer, looks things up in your tools, opens anonymous intake for the room, logs decisions and commitments as they are confirmed, and coaches the person who invited it privately. After the call it emails a recap and queues proposed actions for approval. Side by side Native assistants Mainroom --------- Category Platform-native assistant AI meeting agent Works on Meet, Teams and Zoom One platform each All three, the same agent Appears as a participant with a voice No (text or panel) Yes Summaries and recap Yes Yes, plus a ledger carried to the next meeting Answers questions about the meeting Yes, in a panel Yes, aloud when addressed, or privately on the companion page Shared-screen slides, whiteboard, live demos No Yes Lookups in your own tools mid-call Limited to the vendor's ecosystem Any MCP server plus web search Anonymous intake for the room No Yes Custom persona, manners and tools per agent No Yes Requires a platform license Yes No; joins from a calendar invite Pricing Included with or added to the platform plan Free during beta Which to choose The native assistants are the easiest to turn on if your whole company lives on one platform and what you need is summaries and questions about the transcript. Mainroom is for teams whose meetings span platforms and customers, and who want an agent that does things on the shared screen and in their own tools. The two can coexist: a Mainroom agent is just another participant. A note on accuracy Details about other products on this page come from their public product pages as of September 17, 2026 and may have changed; check the vendor's site for current features and pricing. Mainroom details are current as of the same date. Corrections: hello@mainroom.ai. Q: Can I use a native assistant and Mainroom together? A: Yes. A Mainroom agent is one more participant in the meeting and does not need to be the recorder of record. Teams that like their current notetaker keep it. Q: Does Mainroom produce a transcript and summary too? A: Yes. Every call ends with a recap email matched to the agent's role and a ledger of decisions, commitments, open questions and risks that carries to the next meeting with the same people. Q: Does anything get written to my CRM without approval? A: No. Writes to systems of record arrive as proposed actions that the agent's owner approves or discards. --- # Compare URL: https://mainroom.ai/compare/ Updated: 2026-09-17 Mainroom is an AI meeting agent: a participant that speaks, presents and acts during the call. Most products in this space are AI notetakers that record and summarize, or platform assistants built into one meeting tool. The pages below compare Mainroom with each, using only what the other products' own pages say. The categories at a glance AI notetakers (Otter, Fireflies, Read AI) Platform assistants (Zoom AI Companion, Copilot, Gemini) Mainroom ------------ Joins as a participant Yes No (panel) Yes, with a voice Transcript and summary Yes Yes Yes, plus a carried-forward ledger Speaks when addressed No No Yes Acts on the shared screen (slides, whiteboard, demos) No No Yes Lookups in your own tools mid-call No Vendor ecosystem Any MCP server plus web Works across Meet, Teams and Zoom Yes One platform each Yes Build your own agent No No Yes Writes to CRM and tickets Automatic Limited Proposed, with approval Comparisons - Mainroom vs. Otter.ai — Otter.ai transcribes and summarizes meetings. Mainroom puts an AI participant in the meeting that speaks, presents, draws and acts. - Mainroom vs. Fireflies.ai — Fireflies.ai is a meeting recorder and note-taker with integrations. Mainroom is an AI participant that speaks, presents and acts during the call. - Mainroom vs. Read AI — Read AI adds engagement analytics to meeting notes. Mainroom puts a participant in the meeting and coaches one person privately while it happens. - Mainroom vs. Zoom AI Companion, Microsoft Copilot in Teams, and Gemini in Google Meet — The native assistants live inside one platform and answer questions about the meeting. Mainroom is a cross-platform participant that speaks, presents and acts. Background - What is an AI meeting agent? - AI notetaker vs. AI meeting agent --- # FAQ URL: https://mainroom.ai/faq/ Updated: 2026-09-17 What is Mainroom? Mainroom is a platform for building AI agents that join live meetings on Google Meet, Microsoft Teams and Zoom as participants. An agent has a camera tile and a voice, listens to the whole room, speaks when addressed, presents on the shared screen, draws on a whiteboard, runs live browser demos, looks things up in connected tools, and emails a recap afterwards. You invite one by adding its email address to a calendar invite. Which meeting platforms does Mainroom work with? Google Meet, Microsoft Teams and Zoom. The agent joins as a regular participant and is admitted from the lobby like any guest. For the full experience, enable captions (that is how it knows who is speaking by name) and allow participants to share their screen. How do I invite a Mainroom agent to a meeting? Add the agent's email address to the calendar invite, or forward the invite to it. The agent joins at the start time and appears in the lobby; the host admits it. There is nothing to install on the meeting platform. Is a Mainroom agent an AI notetaker? No. An AI notetaker records and summarizes. A Mainroom agent participates: it answers questions in the call, puts material on the shared screen, runs demos, teaches on a whiteboard, gathers input from the room, and drafts follow-ups. It also produces a recap, but the recap is a by-product rather than the product. Will it interrupt or talk over people? No. It speaks when spoken to, waits for a gap before reporting findings, and stops the moment a person talks over it. Tell it to go on mute and it stays silent until someone asks it to come off mute. You write its meeting manners in plain language, and the customer-facing templates default to silent unless named. Who can invite my agents? You choose per agent: only you, or anyone with your company's email domain, with no Mainroom account required for colleagues. Agents you build stay yours. What is the private companion page? Whoever invited the agent gets a link to my.mainroom.ai the first time it joins a call. During the call it shows live indicators (talk share, questions asked, longest stretch), short private nudges, and a place to ask the agent questions quietly. Nothing on it appears in the meeting. What happens to my meeting data? Agents announce themselves in the roster and platforms show their standard recording notice. The transcript powers the recap email and, if enabled, the agent's memory; memory and recaps are per-agent switches. Credentials for connected tools are stored server-side and never exposed to the meeting. See the privacy policy for details. Can it write to my CRM or ticketing system? Only as a proposal. Anything that changes a system of record (a CRM note, a ticket, a follow-up email) arrives as a proposed action that the agent's owner approves or discards. Nothing is written without approval. What can it connect to? Any tool that exposes an MCP server (Model Context Protocol), plus built-in web search. Connections are configured per agent in the Studio. Can I take it out of a meeting? Tell it to leave and it says goodbye and leaves. You can also remove it from the invite, or have the host remove it from the call like any participant. Everyone in the room can see it is there; it never joins silently. What does Mainroom cost? Mainroom is free while in beta. Build agents, invite them to real meetings, and tell us what you would pay for. Can other AI systems use Mainroom? Yes. Mainroom exposes a public MCP server at mcp.mainroom.ai with tools to describe the product, list agent templates, and send the demo agent into a meeting. This site also publishes llms.txt and llms-full.txt, and every page has a markdown twin. --- # Agent templates URL: https://mainroom.ai/agents/ Updated: 2026-09-17 Every Mainroom agent starts from a template and is then yours to change: name, voice, tools, meeting manners, memory, and who may invite it. These are the templates in the Studio today. Each one joins Google Meet, Microsoft Teams and Zoom from a calendar invite. Customer Success Agent Sits in on every customer call. Remembers the account, tracks commitments both ways, and coaches the account owner privately. Silent in front of the customer unless named. Team Meeting Agent Runs the recurring team meeting's process: confirms decisions, asks for owners and dates as actions land, and brings back last week's open items. Facilitator Keeps a workshop moving: timeboxes, anonymous intake for the quiet half of the room, a nudge when one voice has had the floor too long. Team Coach Watches how the team works together and coaches the leader privately during the call and by email afterwards. Timekeeper Holds the agenda and the clock. Says when a section is over time and what is still left. Delegate Attends on behalf of someone who cannot make it: relays their brief, takes questions for them, and never commits on their behalf. Silent Coach Joins as a plain notetaker, never speaks, and coaches the person who invited it privately during the call. Sends a debrief afterwards. Roleplay Partner Plays a customer, a candidate or a difficult stakeholder for practice, then breaks character and gives feedback. Sales Support Backs the seller on a live call: pulls product facts and pricing when asked, logs commitments, and drafts the follow-up for approval. Pitch Practice Listens to a pitch, asks the questions an investor or a buyer would ask, and scores the answers. Interview Panel Sits on an interview panel with a consistent question set and a rubric, and writes up the scorecard. Brainstorm Partner Generates alternatives when the room runs dry, plays back themes from an anonymous intake, and keeps ideas from being lost. Retro Guide Runs a retrospective: what went well, what did not, what to change, with owners. Risk Review Reads a plan for what could go wrong and asks the uncomfortable questions before the meeting ends. Advisors A small cast of perspectives (finance, customer, engineering) that weigh in when asked. Everything an agent can do - Speak and listen. Joins as a participant with a camera tile and a voice. Answers when addressed, waits for a gap, and never talks over a person. - Slides and pages on the shared screen. Puts up text slides, comparison tables, charts, timelines and small calculators on the meeting's shared screen, and takes them down when asked. - Whiteboard. Teaches on a shared whiteboard it draws on step by step: boxes, arrows, notes, highlights. Questions about the drawing are answered on the drawing. - Live browser demos. Opens a real browser on the shared screen, follows instructions from the room ("click the pricing tab", "scroll down"), and narrates. - Lookups in connected tools and the web. Runs a background search in the team's connected tools (any MCP server) or the web, and reports when there is a gap in conversation. - Generated images. Generates an image from a description and puts it on the shared screen. - Anonymous intake. Opens an anonymous response link for the room, shows a live count, and synthesizes the responses on request. - Meeting chat and private notes. Posts links and recaps to the meeting chat; sends a private note to one person on request. - Meeting ledger. Logs decisions, commitments with owners and dates, open questions and risks the moment they are confirmed, and carries them to the next meeting with the same people. - Proposed actions. Drafts CRM notes, tickets and follow-ups as proposals the owner approves before anything is written to a system of record. - Live coaching. Sends private nudges and answers private questions on a companion page only the inviter can see. - Memory. Remembers people and prior sessions across calls, per agent, as a switch the owner controls. - Post-call email. Emails a role-appropriate recap after the call: a coaching debrief, an account summary, or a team recap. - Floor control by voice. Goes on mute when told to and stays silent until asked to come off mute. Raises a hand instead of interrupting. Build your own Describe the agent you want in plain language in the Studio and the builder drafts it: persona, voice, tools, manners and a recap format. Or start from any template above. --- # About URL: https://mainroom.ai/about/ Updated: 2026-09-17 Mainroom builds AI agents that join live meetings as participants. It started from a simple observation: AI notetakers were in every meeting and doing nothing while they were there. A participant that could answer a question, put a number on the screen, ask for an owner, or draw an explanation would change how the meeting ended, not just how it was recorded. What we have built - Agents that join Google Meet, Microsoft Teams and Zoom from a calendar invite, with a camera tile and a voice. - A Studio for building agents in plain language: persona, voice, tools, meeting manners, memory, recap. - Meeting manners as a first-class feature: when to speak, when to stay silent, how to be muted, how to behave in front of a customer. See what we learned. - A shared screen the agent can use: slides, tables, charts, a whiteboard it draws on, live browser demos. - A private companion page with live coaching for whoever invited the agent. - Proposed actions rather than silent writes: nothing reaches a CRM or a ticket without approval. - A public MCP server so other AI systems can learn about Mainroom and send the demo agent into a meeting. Who Mainroom is built by Taylor Blake. Reach us at hello@mainroom.ai. Status Free during beta, September 2026. Agents can be invited to real meetings today.