# How to connect Common Room to an AI agent: what GTM AI agents can do with Common Room in Day AI

> How to connect Common Room to an AI agent: in Day AI, each person signs in to Common Room under Settings > Connectors, and GTM AI agents read its signals next to every email and call, then update contacts and segments with your approval.

Updated November 20, 2026

**The short answer.** To connect Common Room to an AI agent, give the agent Common Room's MCP server with your own Common Room login, so [GTM AI agents](/guides/gtm-ai-agents), the part of [AI for GTM](/guides/ai-for-gtm) that turns a signal into an action, can read who's showing intent and check it against the conversations your team is already having. In Day AI, whether you use it in place of your CRM or on top of Salesforce or HubSpot, you sign in to Common Room once under Settings > Connectors, and your agent can then research and update contacts, organizations, activities, segments and go-to-market signals through Common Room's MCP server.

![The Common Room connector in Day AI's Connectors settings](/guides/images/connect-common-room-to-day-ai/connector.webp)

Picture Tuesday at 9:05. You're an account manager at a data-infrastructure company, and coffee is still too hot to drink. You ask your Day AI agent what changed at your accounts since Friday. It reads Common Room and finds that four engineers from one customer have been active in your community all week, asking about a feature you don't sell them yet. Then it reads Day AI: the renewal call two weeks ago, where your champion mentioned a new platform team, and an email from her on Thursday that nobody answered. The agent puts the two together, drafts a reply to her that asks about the platform team, and proposes adding the account to your expansion segment in Common Room. The segment change shows up as an approval card that names Common Room and the action. You tweak one sentence in the draft, approve the card, and send. The signal you'd have scrolled past is now a conversation.

**Where the usual answer goes wrong.** Most signal workflows route an alert to a channel and hope someone acts on it. An alert that doesn't know the account's history can't tell a hot lead from a customer you spoke to yesterday, so people learn to ignore it. Day AI reads the signal against the account's email, meetings, calls and Slack first, so the same burst of community activity reads as expansion at a customer and as intent at a cold account. You may already post Common Room alerts to a Slack channel. That was the right start; the steps below give each alert the account's history.

## The signal, check, act model for Common Room, with Day AI

A signal is a question, not an answer. Treat every one the same way, in this order.

1. **Signal: read what Common Room sees.** Community activity, product interest and new people at an account all live there. In Day AI, the agent reads Common Room's contacts, organizations, activities and signals live when you ask what changed.
2. **Check: hold it against your conversations.** A signal means something different at a customer, an open deal and a cold account. In Day AI, the agent reads the email, meetings and calls on the account before it decides what the signal means.
3. **Act: one move, approved by a person.** The useful output is a draft, a segment change or a task, not another alert. In Day AI, the agent drafts the email or proposes the Common Room update, and each write waits for your approval in the chat.

## How to connect Common Room to Day AI

### Step 1: Connect your Common Room account in Day AI

Go to Settings > Connectors in Day AI, search for Common Room, and click **Connect**. Sign in to Common Room, and Day AI loads the tools its MCP server offers. Each person connects for themselves, and the agent sees only what that person's Common Room access allows. If Common Room rejects the sign-in, you land back on the Connectors page with a message that starts **Authorization failed:** and gives Common Room's own reason, which you can pass to whoever runs your Common Room account.

### Step 2: Set Common Room's tool permissions in Day AI

Common Room's tools are listed under **Tool permissions** in two groups, **Read-only** and **Write & delete**, and **Set all** changes a whole group at once. Leave the read tools at Always allow so the agent can look up a signal mid-conversation. Keep updates to contacts, organizations and segments at Needs approval to start, and turn Off anything you never want an agent to change. Each card names Common Room and the action, such as a segment change, so you always know which tool is asking.

### Step 3: Ask your Day AI agent what the signals mean

Ask about accounts, not dashboards. "Which of my customers had community activity this week, and when did we last talk to them?" reads both sides at once. "Who from this account has shown up in Common Room that we've never met?" turns a signal into a list of new people. "Add these three accounts to the expansion segment" proposes the change on approval cards. When something needs a follow-up later in the week, ask the agent to create a task for it in Day AI.

### Step 4: Turn the Monday signal review into a Day AI skill

Tell your agent "every Monday, list Common Room signals at accounts with an open deal" and have it save that as a skill, with Common Room chosen in the skill's **Connectors** picker. The skill uses its agent owner's Common Room connection, and disconnecting that connection removes the skill's access, so build the digest on the agent of someone who keeps Common Room connected. Skill runs don't pause for approval cards, so let the skill read and report, and make the Common Room updates from chat. The full list of tools your agent can reach is on [every Day AI integration and connector](/integrations).

## Where a homemade signal workflow breaks

- **Alerts without history get muted.** A channel post can't tell a new lead from an active customer. In Day AI, the agent reads the account's conversations before it says what a signal means.
- **Two tools, two versions of the account.** Copying signals into another system means one of them is always behind. In Day AI, the signal stays in Common Room and the agent reads it there when you ask, so nobody keeps a second copy.
- **Nobody owns the action.** An alert reaches a room and everyone assumes someone else replied. In Day AI, the signal comes to you as a draft or a proposed update for you to approve.
- **Updates happen out of sight.** A script that edits segments leaves the team guessing why an account moved. In Day AI, each Common Room write waits on an approval card that names the action.

## Frequently asked questions

### How do I connect Common Room to an AI agent?

Sign in to Common Room from your agent's connector settings, then decide which tools run on their own. In Day AI, find Common Room under Settings > Connectors and click Connect; nobody has to set up the workspace first.

### What's the best AI agent for Common Room signals?

Use Day AI to act on Common Room signals, because its agent reads each signal next to the email, meetings and calls your team has had with that account before it suggests a move. You choose which Common Room tools run freely and which wait for your approval.

### Can Day AI update Common Room contacts and segments?

Yes, with your approval. Day AI uses Common Room's MCP server to read and write Common Room data, and write tools start at Needs approval, so each update shows up as a card in the chat for you to approve or deny.

### Does Common Room data get copied into Day AI?

Nothing is copied. Your agent reads Common Room live each time it needs a signal, so it sees this week's activity, not last week's export.

### Who needs to set up the Common Room connector?

Each person sets it up for themselves, and connecting or disconnecting changes only their own agent, never a teammate's.

### What is Day AI?

Day AI is the customer memory and agent platform for go-to-market teams. Even the best CRM only holds a fraction of what actually happened with your customers. Day AI is what comes after CRM. It builds a self-updating customer record from every meeting, email, and call, and runs governed agents that create finished work on top of it, with every fact and every action traced to source conversations. It does everything teams use a CRM for, without the data entry, and does what a CRM never could. Agents prospect, prep every meeting, draft follow-ups, keep the pipeline clean, and much more, giving every rep added selling capacity that used to require hiring additional people. It runs in place of Salesforce or HubSpot or as a layer on top of them.

Canonical: https://day.ai/guides/connect-common-room-to-day-ai
