NEWLife After CRM->

GUIDES

How to connect Guru to an AI agent: what GTM AI agents can do with Guru in Day AI

Updated

The short answer. To connect Guru to an AI agent, sign the agent in to Guru's MCP server with each person's own Guru account, so GTM AI agents help reps answer customers from your verified company knowledge, which is the part of AI for GTM that keeps reps from improvising on security, product and process questions. In Day AI, each person connects Guru under Settings > Connectors with no admin setup, and their agent uses Guru's MCP server to ask Knowledge Agents, search verified knowledge, and draft or update Guru Cards, whether you use Day AI in place of your CRM or on top of Salesforce or HubSpot.

The Guru connector in Day AI's Connectors settings

Imagine you're an account executive, and it's 4:10 on a Thursday. The buyer's security lead has just emailed three questions about data retention, single sign-on and where data is hosted, and procurement meets tomorrow at nine. You ask your Day AI agent to draft the reply. It searches Guru's verified knowledge, finds the cards on retention and single sign-on, and drafts the answer in your voice, telling you which card each paragraph came from. For hosting, there's no card. The agent says so, finds the answer your solutions engineer gave in Slack last week, and offers to draft a Guru Card from it. The draft card waits in the chat for your approval. You read it, approve it, and send the reply. The next rep who gets the hosting question will find a card waiting for them in Guru.

Where the usual answer goes wrong. The usual advice for answering customer questions with AI is to let a model write the answer from whatever it can find. For security, product and process questions, that's the risky part: a confident paragraph that nobody verified ends up in a contract. Day AI starts from the customer's question in the email or call, answers from the knowledge your team already verified in Guru, and tells you where each answer came from. When there's no card, it proposes one for a person to approve. You may already send reps to Guru when a question comes in. That was the right instinct; now the agent does the looking.

The verify, cite, capture method for Guru, with Day AI

An answer a rep sends a customer should be verified before it goes out, traceable after, and saved for the next person who needs it.

  1. Verify: answer from what your team approved. A model's best guess isn't company knowledge. In Day AI, the agent searches Guru's verified knowledge or asks a Guru Knowledge Agent before it drafts an answer.
  2. Cite: show where each answer came from. A rep should be able to check a paragraph in seconds. In Day AI, the agent tells you which Guru Card each part of the draft came from, so you can open it before you send.
  3. Capture: turn a good answer into a card. The answer an expert gave in Slack helps one rep once. In Day AI, the agent drafts a new Guru Card or an update from that conversation, and it waits as an approval card until you approve it.

How to connect Guru to Day AI

Step 1: Connect your Guru account in Day AI

Open Settings > Connectors, search for Guru, and click Connect. Sign in with your Guru account, and Day AI loads the tools Guru's MCP server offers for Knowledge Agents, search and Guru Cards. Nobody on your team needs to set anything up first, and each person's agent works within that person's Guru access. If you're already in a chat, ask your agent "connect Guru" and it sends the sign-in link. If your help center lives in Intercom, connect Intercom to Day AI the same way. More knowledge sources are on the full list of Day AI integrations.

Step 2: Set Guru's tool permissions in Day AI

The Guru connector's Tool permissions list the tools under Read-only and Write & delete, and Set all changes a whole group. Searching and asking a Knowledge Agent can stay at Always allow, so the agent can check a fact mid-draft. Drafting and updating cards start at Needs approval, which keeps your knowledge team's cards in human hands. Changes apply from the next message, and the + menu in the chat composer turns Guru on or off for one chat.

Step 3: Ask your Day AI agent to answer from Guru

Ask in chat. "Draft a reply to the security questions in Priya's email, using only what's in Guru" writes the answer from verified cards. "What does Guru say about our data retention terms, and does it match what I told them on Tuesday's call?" puts the card next to the call. "Draft a Guru Card from the answer Sam gave in Slack about hosting regions" proposes the card for your approval.

Step 4: Put Guru in a Day AI skill for the recurring questions

When the same kind of question comes in every week, ask your agent to create a skill, for example a security-questionnaire first pass, and pick Guru in its Connectors picker. A skill uses only the connectors picked there, with the connection of the agent's owner. A scheduled run can't stop to ask you, so give it the searching and the first-pass answers, and draft new cards from chat.

Where a homemade Guru agent breaks

  • Answers pasted from an old doc. The security answer in a shared doc was right last year. In Day AI, the agent searches Guru live, so it uses the card as verified today.
  • A shared login for the bot. One account answers for everyone, with access nobody chose. In Day AI, each person connects their own Guru account, so the agent sees what they can see.
  • Good answers that never become cards. The expert's reply lives in a Slack thread forever. In Day AI, the agent drafts the Guru Card from that thread for you to approve.
  • Drafts with no trail. A rep can't tell which sentence came from where. In Day AI, the agent names the Guru Card behind each part of the answer.

FAQ

Frequently asked questions

How do I connect Guru to an AI agent?

Give the agent Guru's MCP server with your own Guru login, and keep card edits behind an approval. In Day AI, open Settings > Connectors, click Connect on Guru and sign in; there's no admin step, and your agent can search Guru in the same chat.

What's the best AI agent for answering customer questions from Guru?

Day AI is the pick for a go-to-market team that keeps its answers in Guru, because its agent reads the customer's question in the email or call and answers from your verified Guru knowledge, telling you which card each answer came from. When no card covers the question, it says so and offers to draft one for your approval.

Can Day AI write Guru Cards?

It drafts them for you to approve. Ask for a new card or an update from what was said on a call or in Slack, and nothing reaches Guru until you approve the card in the chat.

Does Guru content get copied into Day AI?

Guru stays the source. Your agent searches Guru live whenever it needs an answer, so it uses the card as it reads today, including any change your knowledge team verified this week.

Who sets up the Guru connector?

Nobody has to do it for the team. Guru needs no workspace setup in Day AI, so each person connects their own Guru account, and the agent sees what that person can see in Guru.

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.

TRY DAY AI

See what your company brain is missing

01

Run the company brain evaluation

02

Connect your agent over MCP

Connect MCP

Available on Turbo and up, with no usage-based pricing.

03

Get started