AI Agents on monday.com

 

 

Note: monday AI Agents are currently in a gradual release. If you don't have access yet, it will be available to you soon.
Who can use this feature
gray-cloud.svgAvailable on monday AI platform, availability on all products coming soon

Use this guide to understand how monday agents work, when to use them, and how to get started with the monday AI agent builder.

 

What are monday agents?

monday agents are AI agents built inside monday.com that act within the context of your work. They monitor activity, make decisions based on your defined rules and priorities, and execute tasks end to end, all directly inside your boards and workflows.

Unlike rule-based automations, agents can analyze information, apply judgment based on your defined priorities, and take the next best action, all within the boundaries you set. Agents operate within your permissions and guardrails, so you stay in control of what they do, where they act, and how they communicate.

monday agents include:

  • Expert Agents – Pre-defined agents designed for common end-to-end use cases. These include the CRM AI Lead Agent and the CRM Sales Agent
  • monday AI agent builder – A flexible builder that allows you to create custom agents tailored to your workflows.

 

What can agents do?

Custom agents built with the monday AI agent builder can:

  • Understand context – Use your boards, data, docs, workflows, and permissions to analyze and connect signals. Agents can also understand external context, by connecting external files with specific knowledge.
  • Make decisions – Apply rules, priorities, tone, and thresholds you define to triage, route, escalate, and prioritize work.
  • Take action – Create and update items, assign owners, change statuses, draft messages, log outcomes, and execute follow-ups.
  • Operate within governance controls – Act only where permissions allow and within defined access boundaries.
  • Run based on triggers – Execute when specific events occur, at defined times, or on an ongoing cadence. Agents will keep working even when you are asleep, taking care of your work for you even when you are unavailable to do it yourself.

Agents can support work such as capturing and triaging incoming requests, researching and summarizing information, routing tasks to the right owner, drafting follow-ups, maintaining records, and coordinating work across teams.

 

Billing and credit consumption

Attention: Credit consumption for monday agents starts on June 8, 2026, for Pro plans and below. Enterprise plans are currently exempt, and will transition to credit-based usage at a later date.

 

Credit consumption depends on the complexity and depth of each task your agent performs. For full details, see the AI Models and Credits: understanding and optimizing consumption article.

 

How to get started with the monday AI agent builder

To create your first agent:

First, click on the AI Agents tab in the left navigation pane. Next, click on the + New agents tab on the left pane menu. 

You can describe your ideal agent in plain text, choose a pre-defined starting prompt, or click on + Start from blank at the top right of the page

You can also use pre-built agents in Agents made for you. These templates are personalized and specifically generated for your role within your company and based on the data on your boards.
 

 

Note: You can also connect agents built on external platforms into monday.com using Bring Your Own Agent. For details, see the Bring your external agent into monday.com article.

 

Next, you will see the following page. Here you can customize your agent as you wish. To continue, click on Onboard:

 

Note: Once onboarded, your agent will be automatically activated with the instructions provided. You can continue chatting with it and refine its instructions, adjust tools, permissions, and triggers.

 

To pause your agent, click the Pause agent button, located in the top right three-button menu of the Agent builder. In a Paused state, you can continue building and editing, but the agent cannot run. 

To resume, take your agent live again by clicking the Go live button located in the This agent is paused message at the top of the Agent builder

 

 

Customizing and managing your agent

There are several components in the Agent builder that give you the ability to customize, tweak, and manage your agent. All changes you make, whether through the chat or the config panel, are saved and applied automatically.

 

Agent left panel chat

Use the agent chat on the left side panel to ask the agent builder about your agent's abilities, troubleshooting questions, make any changes you need, and upload any files or boards as context until you are satisfied with your final result. 

The agent chat will guide you through the entire building experience, suggesting improvements, additional context to add, and when you can run a simulation to test out the agent. Not sure what can be done? Ask the agent what it can and cannot do directly!

It will point out to you what is missing, what you can add to your agents, and will let you know about any missing information. If you have any questions, you can of course ask, but you do not need to decide on your own how to best optimize your agent. If you are unsure, the chat is there to help you.

 

After the agent performs any configuration action (such as adding an instruction, trigger, or tool), an Undo option appears immediately. The agent also supports full editing and removal of its own configuration. 
 

Note: Undo covers configuration actions only, not the content the agent produced (such as an update it wrote on an item).

 

 

Brain tab

Use the Brain tab to choose triggers, fine-tune instructions that have been generated by the chat, select tools and access permissions, add connections to integrations, add context via files and documents, and much more to bring your agent to life.
 


Instructions

You can review the instructions the Agent chat has generated for your agent. Make any additions or edits needed directly in the box, or click on Expand to open up the full document to make edits.

Remember that the agent cannot perform actions outside of what it is explicitly permitted to do. Proper configuration will ensure that your agent runs safely and in a predictable manner.


 


Knowledge and access

Use the Knowledge and access options to grant access to monday boards and other files, increasing context awareness and improving results.

You can also specify the exact permissions you want to grant to the agent, for example whether the agent can only read the board data or if the agent can also edit and add to the board data. You have full control over the permissions granted to the agent, allowing you to be as private or open with the agent as you prefer.

If there is additional context (for example, FAQs, governance and legal information, etc.) that's located outside of your monday boards that is relevant for your agent to use, you can upload files within this section. Supported file types are: pdf, doc, txt, excel, and csv. 

 

The agent proactively recommends relevant boards to connect based on your instructions. Recommendations appear in the Knowledge and access section and via the agent chat, with up to 5 suggestions at a time.

 

Tools

Use the Tools section to maximize your agent's abilities with web search capabilities and connections to platform integrations.

Agents can also connect to external tools through integrations, allowing them to pull in information, update records, and keep systems aligned without manual copy and paste. 

To add a tool, describe what you need in the agent chat. The agent will:

  • Identify the right tools based on your request
  • Ask the questions needed to resolve any ambiguity
  • Surface the connection step directly in chat
  • Confirm exactly what was configured so you can review before moving on

 

Tools are classified as Read (search, pull data) or Write (send, create, edit, delete). Read tools are active immediately after connection. Write tools are added as inactive and the agent will prompt you to review the setup. You can explicitly activate the tool when you're ready. 

Guardrails define the boundaries of what a tool is and isn't allowed to do. For example, you can restrict the agent to only post in a specific Slack channel, or only email certain contacts. The agent asks these questions as part of the setup flow and applies your answers across all relevant tools at once, so you don't need to repeat yourself for each one.

To use dynamic values in your connections, make sure you define the parameters explicitly in the agent instructions in the Brain tab, or via chat. 

Supported integrations include email providers such as Gmail and Outlook, messaging platforms, calendars, CRMs, developer tools, file management tools such as Google Drive, and more. 
The full list of available tools is always up to date in the Brain tab under the Tools dropdown.

 

 

Check out some of the available tools
  • Outlook
  • Gmail
  • Slack
  • MS Teams
  • Outlook Calendar
  • Google Calendar
  • Salesforce
  • HubSpot
  • Jira Data Center
  • GitHub
  • Confluence
  • Xero
  • Google Drive
  • Box
  • Calendly
  • Twilio
  • Trello
  • GitLab
  • Active Directory
  • JotForm
  • Stripe
  • PagerDuty
  • Pipedrive
  • Eventbrite
  • Typeform
  • Basecamp
  • Bitbucket
  • Resend
  • Harvest
  • Copper
  • Pingdom
  • Toggl
  • Todoist
  • Telegram
  • Zoom

 

Note: Some tools connect via MCP (Model Context Protocol), exposing a wide range of capabilities through a single connection. The individual capabilities aren't displayed in the UI. You can ask the agent what a tool can do, and it will provide detailed information.


Here you can also:

  • Enable web search - toggle access on and off using the switcher to allow your agent access to searching the web for more accurate search results
  • Grant access to the AI Notetaker - the agent will have access to the meeting records, summaries, transcripts, and action items. 
  • Add a custom MCP - In addition to the built-in integrations, you can extend your agent's capabilities by connecting any external service that exposes an MCP-compatible endpoint. This allows you to go beyond the built-in toolset and connect services you control directly to your agent. Under tools, click on + Add custom MCP. In the following page, add the URL of your MCP server to connect your custom MCP. Then give it a name and add your Access token/API key. You'll then see your MCP listed under your tool list.

 

Skills

Add skills that give your agents the ability to complete specific tasks.

Click on + Add skills. From there, choose a skill from the list that fits your needs, then click Add skill. You can also chat with your agent to create a skill together. 

 

 

Choose your LLM model

Under Model you can choose the primary AI model for your agent. Additional models may be used by the agent to optimize performance. 

 

 


Jobs tab

Jobs let you break your agent's work into separate, focused tasks. Each has its own trigger and its own scoped instructions. Instead of one global instruction set, each Job handles a specific piece of work, making your agent more precise and efficient.

Each Job contains:

  • A trigger that defines when it runs
  • Instructions scoped specifically to that task

Jobs can run in parallel, meaning a complex workflow can be split into multiple Jobs that execute simultaneously. Jobs can also be scheduled to run on a recurring basis: daily, weekly, or monthly.


There are two ways to create Jobs:

  • Agent-suggested, via chat - The agent identifies that a task is recurring or complex and proactively suggests creating a Job. You confirm, and the Job is created.
  • Manual, from the Jobs tab - Go to the Jobs tab, create a Job directly: pick a trigger, write instructions specific to that Job, and give it a name.

Two default Jobs exist on every agent: When Assigned and When Mentioned. These are always available regardless of how many additional Jobs you create.
You can ask the agent to perform a one-off task at any time, regardless of its existing Jobs. The agent executes it and returns the result without running its full instruction set.

 

Note: If you had triggers configured before Jobs were introduced, they have been automatically carried into Jobs. Your agent continues running with the same behavior, no action is required.

 

Channels tab

Here you can define how your agent communicates with people in external apps. These apps include:

  • WhatsApp
  • Telegram

With this feature, you can directly activate and also communicate with your agent from outside the monday platform using the communication tool you prefer.

For example, if you're on the go and prefer communicating on WhatsApp, simply connect your agent to WhatsApp and message them directly on your WhatsApp mobile app!

 

Activity tab

Your agent is active as soon as it's created; there is no manual activation step. Once the triggers and rules you configured are met, your agent will run. 

In the Activity tab, you can see all of the runs your agent has completed, all in one place. You can monitor the status of each run and always be on top of what may require your attention. You can see the date of the run, which apps it used, and how many AI credits were consumed, if any.

Click on a particular run in order to view more details of what has been done or what else needs to be added in order for the agent to get the work done:

 

If a run fails, you will receive a bell notification, even if you are not viewing the agent at the time. When you open a failed or partial run, a What Went Wrong panel shows what happened, why, and a suggested fix. The run summary also includes details on what the agent was attempting when a tool failed, giving you context beyond a simple failure message.

 

Build a custom tool with Tools on Demand

If an integration you need is not available inside the platform, you can build one directly from the agent builder chat. Describe the external service and the actions you need in plain language, and the tool will be researched, generated, and deployed to your account automatically.

 

Note: Tools on Demand is currently in Beta.

 

Once built, the tool is available to any agent on your account. You can find it under the Created by Me tab when adding tools to an agent.

 

To build a tool, type your request into the agent builder chat. For example:

  • "I need to fetch open issues from Linear"
  • "Build a Stripe integration to create and list payments"
  • "I want to manage contacts in HubSpot"

 

If no existing integration matches your request, the agent will offer to build one. Click Yes, build it to start. The build runs in the background and typically takes a few minutes. You can continue working on your agent while it runs.

When the build completes, a connection card appears in the chat. Click Connect and enter your API key or token for the external service. You can find your API key in the Settings or Developer section of the external service.

 

Tools on Demand limitations
  • Only API key or token-based authentication is supported. Services that require OAuth-only login cannot be built at this time.
  • Only action-based operations are supported (create, read, update, delete). Triggers and webhooks are not supported.
  • As this feature is in Beta, some builds may fail. If a build fails, try rephrasing your request or simplifying the actions you need.
  • Editing a tool after it has been built is not yet supported.

 

Control, permissions, and governance

monday agents are designed with enterprise-grade controls in mind. This ensures agents operate predictably and in alignment with your organization's structure and policies.

  • Agents only access what you are allowed to access. You can narrow this access to the minimum that's necessary for the agent to run. For example, you can grant the agent permission to read only particular boards or the same viewing permissions as a particular user is allowed to see.
  • Creators control where agents can act (specific boards, spaces, teams). Creators are also able to decide what the agent is allowed to do within a specific scope. For example, the agent can only view a particular board or can edit only a specific board.
  • You can review a transparent log of actions and reasoning in the Activity tab.
  • Agents can be paused at any time. In a Paused state, you can continue building and editing your agent, but it cannot run. To resume, take your agent live again.
  • Agent owners can assign editors and viewers, but currently, ownership cannot be transferred.

To learn more about agent governance, please see the AI permissions and governance article.

 

Share your agent

Note: The ability to share agents is currently in a gradual release. If you don't see this option for your agent yet, it will be available to you soon.


Once your agent is created, you can share it with other members so they can interact with it directly, without needing to build or configure anything themselves.

To share your agent, click the Invite button at the top right of the agent builder. Search for and select the users you'd like to invite, and they will be added to the agent.

Shared users can trigger the agent, chat with it, and view its settings in the Brain tab. They cannot edit the agent's instructions, tools, permissions, or any other configuration.
 

How sharing works

When you share an agent, it retains its existing context and permissions. Shared users interact with the agent using the full context it already has access to, if the users have the same access.

  • The owner grants context and permissions to the agent during setup.
  • When the agent is shared, authorized users can invoke it with that full context.
  • If the agent has access to a private board or doc that a shared team member cannot access, an access request is sent on behalf of the agent. The request includes all users who will have access to that board through the shared agent, so access can be granted at once.
  • Coming soon: you will be able to invite users as "Editors" for your agent. 

 

Note: Agents with external tools configured (such as email sending integrations) and Notetaker cannot be shared at this stage.

 

Other home page features

Additionally, on the agent builder home page, you can view any pre-existing agents. Add new capabilities to existing agents by visiting the Skill Hub:

 

Browse through existing skills, upload skills, or create new ones by clicking on Create skills:

 

After you click on Create skill, you will be able to chat with Sidekick and describe exactly which skill you want your agent to perform and customize it as you wish. You can also manually write out what you'd like to create. All changes are saved automatically.

 

Post-build agent experience

What can I do with an agent in a board?

  • Assign the agent to items to trigger it. 
  • Open an existing agent for editing / create a new agent from the agent button at the top-right pane: 
  • Use Sidekick to interact and improve the agent
  • View its actions and undo from the activity log:
  • Follow your agent's activity on the board - while the agent works on your board, you can see it happen in real time. The agent's avatar moves between cells like a collaborator. 
    To enable this on your board, enable the Live agent mode toggle in the Agents button at the top right of your board. 

 

What can an agent do within a monday doc?

  • Pull context or reference information from it when generating status updates
  • Use it as a knowledge base for project standards or guidelines
  • Include relevant sections in the agent's reports
  • Write to a doc (like adding status updates directly to it), by updating its permissions to read-write.

 

Artifacts

The Artifacts tab centralizes everything your agent has produced, including images, HTML outputs, monday docs, and monday boards. All outputs are stored as named, searchable artifacts accessible across conversations.

 

Agent memory

Agents retain memory across runs and can reference past work and outputs, persisting across conversations. Your agent can remember facts about you, such as your preferences and working style, as well as behavioral rules you set, like "always ask before sending an email". 
 

You can access the agent's memory by clicking on Manage memory located in the top right three-button menu of the Agent builder. The memory functionality can also be fully switched off if you prefer a fresh start each session.

You can also ask your agent to remember or forget anything at any time. 

 

Troubleshooting

Agents continuously evaluate and act based on the provided instructions, meaning that they run in a continuous loop.

If the agent performs unexpected actions outside of your definition scope or appears to be stuck in a loop:

  • Use the abort / stop button to stop the run
  • Revisit and refine instructions in the Brain tab to ensure they detail exactly what you want your agent to accomplish
  • If everything is defined correctly but the behavior persists, please reach out to support for help

Please note that any changes or updates made while an agent is running will not affect the current run and will only apply to subsequent runs.

Before escalating an issue, you can usually investigate agent behavior directly in the product. The agent setup is shown in a readable format, so you can review its instructions, permissions, tools, and connected resources without needing technical expertise.

You can also open the Activity tab to review each run, including what the agent did, when it ran, and why it took a particular action.

 

How can you troubleshoot an AI agent?
  • Review the agent's configuration in the UI to understand its permissions, tools, and prompts
  • Go to the Activity tab and open the run log
  • Review the agent's thought process, plan, and actions
  • Confirm triggers, conditions, and inputs are correctly defined
  • Ensure relevant boards, docs, and integrations are connected and accessible
Why can't the agent access a board or data?
  • Confirm the agent has been granted access to the relevant boards or docs
  • Verify the creator of the agent has access to those resources
  • Check if any recent permission changes were made
  • Reconnect or reconfigure the resource in the agent setup if needed
Why did the agent perform an action it shouldn't?
  • Review the agent's configured actions and logic
  • Check if permissions allow broader actions than intended
  • Validate triggers and conditions that led to the action
  • Adjust configuration to restrict unwanted behavior
  • Run the agent in Preview Mode to validate behavior (will be relevant when preview tab is restored)
Why didn't the agent perform an action it should?
  • Review triggers and ensure conditions are correctly defined
  • Confirm the triggering event actually occurred
  • Check for missing inputs or required fields
  • Validate setup
Why are the agent's results incomplete or incorrect?
  • Review the instructions or prompt given to the agent
  • Ensure relevant data sources are connected
  • Check for ambiguous or conflicting instructions
  • Test with a simplified input to isolate the issue
Why is the agent not triggering?
  • Confirm trigger conditions are correctly defined
  • Check that the triggering event actually occurred
  • Verify the agent is not paused
  • Review the Activity tab for attempted runs
Why can't you access or run the agent?
  • Verify you have permission to access the agent (viewer or editor)
  • Confirm the agent has been shared with you
  • Check workspace or account-level restrictions
  • Re-share the agent if needed
Why are the agent's actions unclear or unexpected?
  • Go to the Activity tab
  • Review what the agent did, when, and why
  • Cross-check with the configured logic
  • Adjust configuration if behavior is not aligned
What should you do if you have concerns about data security or access?
  • Ask clarifying questions to understand the concern (such as data exposure, permissions, or integrations)
  • Review which boards, docs, and integrations are connected
  • Confirm the agent operates within existing and defined permissions
  • Restrict access to sensitive resources if needed
Why does the agent work in preview but fail live?
  • Compare Preview Mode versus live conditions (data and triggers)
  • Check if live data meets trigger conditions
  • Verify permissions in the live environment
  • Re-run tests with real scenario inputs
Why were credits charged for an agent run?
  • Review what actions the agent performed during the run
  • Check which features or tools were used
  • Correlate usage with credit consumption logic
  • Set expectations for how similar runs may be charged going forward
Why can't the agent be edited or managed?
  • Confirm you have editor permissions
  • Identify the current owner of the agent
  • Note that ownership cannot be transferred
Why is the agent triggering itself in a loop?

Agents are blocked from triggering themselves to prevent unintentional credit consumption — for example, a rule like "every time an item is created, create an item." When a self-trigger is blocked, an explanation appears in the Activity tab along with a notification containing fix instructions. Review and update your trigger configuration to resolve the loop.

 

Tip: In many cases, the fastest way to troubleshoot a monday agent is to compare the configured instructions and permissions in the Brain tab with the run details in the Activity tab.

 

How are monday agents different from automations, monday sidekick, and Agent Factory?

AI Agents vs. Automations

Automations follow predefined "if this, then that" logic.

monday agents go further by generating content, evaluating context, and choosing actions dynamically, while still operating inside the structure and guardrails you define.

This allows agents to handle more complex scenarios, ambiguity, and decision-making across multiple steps.

 

AI Agents vs. monday sidekick

monday sidekick is designed for conversational support and on-demand assistance.

monday agents are built for execution inside monday.com. They run processes end to end at scale, monitor activity continuously, and take action based on triggers and defined workflows.

monday agents can run anytime, whether you are working or not. They can work even as you sleep, so you can maximize your workflow even when you're not actively working yourself.

 

AI Agent Builder vs. Agent Factory

monday agents are a capability built directly into monday.com.

Agent Factory is a standalone product outside of monday.com designed for independent agent creation.

 

Other resources

Agent templates

Use the AI Agent template center to jump start the agent building process by starting out with conveniently pre-made agent templates.

 

Agents in mobile

You can also use your agent on the go! For more information, please check out the Sidekick mobile app article.

 

FAQs

Do I need admin permissions to create or manage agents?

Yes. Admin permissions are required to create or manage monday agents.

Can I pause my agent?

Yes. You can pause your agent at any time. In a Paused state, you can continue building and editing, but the agent cannot run. To resume, take your agent live again.

Can agents work together in one flow?

Yes. With AI workflows, multiple agents can operate in a single flow, each handling a different stage of execution.

How are agents different from automations?

Automations follow fixed rules. Agents evaluate context, generate content, and choose actions dynamically within defined guardrails.

Where can I troubleshoot a monday agent run?

Open the Activity tab to review what the monday agent did, when it ran, and why it took specific actions.

If you have any questions, please reach out to our team right here. We’re available 24/7 and happy to help.

Last modified on

Still have questions?
We can help.

Chat support

Get in touch with our support team through our chat for any questions, concerns, or inquiries.

Community forum

Learn, share ideas and connect with other monday.com customers.

Expert help

Hire a monday.com expert to optimize your workflows.

Is this article helpful?
Help us improve our articles.