Who can use this feature

Available on monday service

Service AI agents are AI teammates inside monday service that answer requests, route conversations to the right agent, and hand conversations to a human expert when needed. They give requesters faster answers and let your experts focus on the requests that need them. This article is your starting point: it shows how Service AI works end to end, outlines the setup, and walks you through creating a team and configuring each agent.

What are Service AI agents?

Service AI agents are the AI teammates you create and manage across your organization. Instead of one general agent, Service AI uses multiple specialized agent teams. Each agent is set up for a specific type of request, such as benefits, payroll, onboarding, or IT support. This improves routing accuracy and keeps your setup easy to maintain.

How Service AI handles a request

A requester sends a request through a channel: portal AI chat, email, a ticket form, or Microsoft Teams. monday service creates a ticket on the connected Tickets board. The Service AI Supervisor matches the request to the agent whose topics fit best, and that agent replies using the knowledge you gave it.

If no agent matches, knowledge coverage is too low, or the requester asks for a person, a human expert takes over. To see how the Supervisor makes these decisions, see Test and monitor monday service AI.

Set up Service AI

Setup has four stages. This article covers the first, and two companion articles cover the rest.

1Create a team and configure its agents. Define each agent's topics, knowledge, and playbooks. The steps are below.

2Add knowledge. Connect docs, files, web pages, Confluence Cloud, or SharePoint so your agents have answers to draw from. See Set up knowledge and channels for monday service AI.

3Connect channels and your portal. Choose how requests come in and set up AI chat in your Customer Portal. See Set up knowledge and channels for monday service AI. For channel types, setup steps, and channel health, see Channels on monday service AI. For portal design and access, see Customer portal settings on monday service.

4Test and monitor. Test the full flow before requesters use it, then track how your agents perform. See Test and monitor monday service AI.

Open Service AI

1Open monday service.

The left pane includes Requests, where you access your workspaces and Tickets boards, alongside the tabs that make up Service AI:

• Agents: build and manage your AI teams
• Performance: monitor how your Service AI agents are doing
• Portals: manage where requesters go for self-service and AI chat
• Knowledge: manage docs and connected knowledge sources
• Channels: manage how requests come in

2Select Agents to get started. On the Service agents tab, your agent teams are under Service Frontliners, and the Supervisor is under Service Ops.

Create a team

To create a team, select New team at the top right of Service Frontliners. Choose the team type that fits the requests you want the team to handle, and the requests workspace where your team collaborates. The requests workspace needs at least one Tickets board linked to the team.

Your new team appears under Service Frontliners alongside your other teams. Each team shows its name, its requests workspace, and icons for its connected channels. For example, an IT team may include agents such as IT Support, Access & Identity, and Device & Asset agents. Each agent shows its topics covered and its status (Active or Inactive). Once an agent starts resolving requests, its resolution rate appears next to the status. To find an agent, search by name or type.

Select + New agent on a team to create additional agents beyond the presets. Use the three-dot menu next to it to add another agent, open team settings, or delete the team. Team settings let you change the connected requests workspace and view or add incoming request channels.

Next to + New agent, icons show which channels are connected to the team's requests (Portal AI chat, email, and forms). Hover over an icon to see its connection status. To manage channels, see Set up knowledge and channels for monday service AI.

Configure an agent

Select an agent to open its setup panel. The left panel has two groups. Under Agent's settings: Identity, Knowledge, Playbooks, and Requester context. Under Agent's performance: Coverage test and Activity. You can also activate or deactivate, or delete the agent from the top of the panel. An agent must be active in order to answer requests. To activate an agent, it must have both topics covered and knowledge.

Identity

The Identity tab shows the agent's team, its name, a short description, and who created it. You can rename the agent with the pencil icon next to its name, and choose an avatar. Under Automatically handles tickets, On these boards and From these channels show the Tickets boards and channels the agent handles tickets from.

Topics covered is required, and it is the main thing that affects routing. The Service AI Supervisor uses it to decide which requests to route to that agent. It does not use the knowledge base to make that routing decision, so your topics should be as specific as possible.

Only include capabilities your team supports. For example, if you do not use device-management tools, leave asset and device-management help out of the topics. Tone of voice and answer length are set in the Playbooks tab.

Knowledge

In the Knowledge tab, add the information the agent uses to answer requests. Knowledge comes in two types. Sourced knowledge is content you want requesters to see, and the agent cites the source and includes a link when it answers. Unsourced knowledge is shared without showing its source to the requester.

Select Add sourced knowledge to add a source from:

• monday docs: pull in a monday doc.
• Add files: upload a file directly. See file limitations below.
• Web page: sync a public web page. The agent syncs the page and one level of linked sub-pages, and refreshes weekly. Pages that require login aren't supported.
• Confluence Cloud Spaces: pull from a Confluence Cloud space your admin has connected under Knowledge > Connectors. You can select specific spaces, or a specific hierarchy of articles within a space, for this agent. Content syncs from Confluence Cloud once a day.
• SharePoint Sites: pull from a SharePoint site your admin has connected under Knowledge > Connectors. You can select specific sites, or a specific folder hierarchy within a site, for this agent. Content syncs from SharePoint once a day.

To connect Confluence Cloud or SharePoint, see Set up knowledge and channels for monday service AI.

Select Add knowledge under Unsourced knowledge to add a source from the same list, plus monday board. A monday board can be any board across your account, except Tickets boards. File columns aren't extracted, and board content isn't matched against personal information, so use caution when adding boards that contain sensitive data.

File upload limitations

Playbooks

In the Playbooks tab, tell the agent how to behave. Communication sets how the agent talks with people, with rules such as Answer length, Tone of voice, and Personalized greeting. Human handoff decides when and how the agent escalates to a human agent.

Playbooks guide how the agent communicates, answers, and hands off. They don't change how it works with tickets. The agent still replies in your connected service channels and follows the same ticket lifecycle. Each rule is one of three types:

• Locked: you can't edit or turn it off, because it uses built-in guardrails. For example, Escalation & handoff and Form submission summary.
• Editable: a system default you can adapt or turn on or off. For example, Hand off automatic replies sends out-of-office, no-reply, bulk, and billing or system emails to a human agent without replying to the requester.
• Custom: a rule you write with Add. It shows its name and who edited it.

Core rules always take priority. If a custom rule contradicts a core rule, monday service blocks it instead of combining the two.

To write a rule, select Add, or select an editable rule to change it. Write the rule in plain language, then save it. monday service checks that the rule is valid and doesn't conflict with core rules. To apply a rule only in some cases, put the condition in the rule text. For example: "If the email subject starts with 'approved', route to a human agent." There is no separate trigger setting.

Escalation & handoff routes the conversation to a human agent when the requester asks for one, after 2 to 3 unsuccessful attempts to resolve the issue, when the issue falls outside the agent's topics, or when the knowledge search returns no results or only irrelevant ones. When a playbook rule causes a handoff, the activity log shows Playbook rule as the reason.

Playbook limits

• Up to 20 custom playbooks per agent or Supervisor, and up to 1,500 characters per playbook.
• Playbooks mainly guide how the agent behaves. Human handoff rules, and Info collection on the Supervisor, are the exceptions, because they trigger actions.
• A rule can only use information the agent has. For example, rules based on the email sender, To, or CC work because that data is passed to the agent.
• Playbook text is part of what the agent reads, so wording from a rule can occasionally appear in a reply. This is reduced, not eliminated.
• You can't see which playbook rules applied to a ticket, except for handoffs, where the activity log shows Playbook rule.

The Supervisor has its own playbooks, including Info collection. To learn how they work, see Test and monitor monday service AI.

Requester context

In the Requester context tab, give the agent context about each requester so it can provide more relevant, specific answers from its knowledge sources. Select Add a board to connect a board that holds requester information, such as vacation balances. The Connected sources table lists each board and how many of its columns are connected.

Coverage test

In the Coverage test tab, review auto-generated questions based on the agent's defined topics and knowledge. The tab shows the coverage level, such as Basic. Each question is marked Covered or Didn't pass. Select a question to see the agent's reply and its reasoning, such as Request in topics and Knowledge sufficient.

Select Regenerate for a new set of questions, or Test within team to test the agent alongside the rest of its team. To test the agent alongside your whole setup, see Test and monitor monday service AI.

Activity

Once the agent is live and handling requests, use the Activity tab to review its activity log. Each entry shows what happened (such as agent replied, ticket resolved, or handed off to human), the related ticket, the channel the request came through, the reasoning behind the action (such as "Answered with KB" or "Matched - high confidence"), and when it happened. Filter the log by event type or time range (last 7 or 30 days).

To see how the Supervisor routes requests across all your agents, see Test and monitor monday service AI.

FAQs

Can I customize how a Service AI agent communicates and hands off in monday service?

Yes. In the Playbooks tab, edit the editable rules or add custom rules. Locked rules can't be changed, and core rules always take priority.

How does monday service route a request to the right AI agent?

The Service AI Supervisor matches the request to the agent whose topics fit best. If no agent matches or knowledge coverage is too low, a human expert takes over.

What do I need before Service AI agents can answer requests in monday service?

You need a team with at least one active agent that has topics covered and knowledge, and a channel connected to a Tickets board.

Which channels work with Service AI agents in monday service?

Portal AI chat, email channels, ticket forms, and Microsoft Teams.

What knowledge can I add to a Service AI agent in monday service?

monday docs, uploaded files, public web pages, Confluence Cloud Spaces, and SharePoint Sites. Unsourced knowledge also accepts monday boards, except Tickets boards.