With Agent Factory, you can create your very own personalized AI agents that perform for you.

Agent Factory is a standalone product from monday.com that is accessible from outside of monday.com and allows anyone to design, launch, and manage intelligent AI agents. These agents are not simple bots; they are powerful tools that can be customized to handle complex workflows.

What is Agent Factory?

Agent Factory was created to scale your business by building specialized AI agents to perform tedious tasks so you can focus on strategic work.

Agent Factory gives you the tools to build a dedicated agent to handle tasks and workflows. You can customize pre-made agents or create a new agent from scratch using a prompt.

The agents can handle everything from a morning brief call to a competitor analysis report, freeing up your time and scaling your business.

How to set up Agent Factory

Log into Agent Factory by clicking here in order to start creating your agent.

Creating your agent

After logging in, you will see the account's existing Agents in My Digital Workforce. Click on + Create agent to get started.

Use either the available pre-made agents or start typing in the chat to explain what you want your agent to do. Include the exact actions you want them to take, resources to use, tone and style of speaking, and any other important notes.

For example, you can instruct your agent to identify themselves in the phone call as being an AI Agent.

Continue reviewing your agent in the Instructions tab and tweaking the prompt in the chat until you're satisfied with your agent's setup.

Once you give instructions, you will see a preview of your agent. You can update as needed with additional prompts:

Customizing your agent

Now it's time to customize your agent according to your needs. You will see all of the various tabs on the left pane menu that will allow you to create an agent tailored exactly for you.

Phone

Use the Phone tab to customize how you want to agent to speak, from greeting to phone number, to voice and accent, speech speed, and more.

Capabilities

Use the Capabilities tab to decide what the agent uses in order to communicate. You can use Skills like monday.com, Gmail, Google Calendar, and Slack, as well as SMS messages. Add more skills by clicking the + Add capability button on the right.

Please note that when using Slack, you can only send messages to public channels or to the channels you have connected to monday.com.

Inputs

You can use the Inputs tab to provide the agent with the relevant variables that it needs to carry out the task, for example name and phone number.

Add as many as you need by clicking on the + Add input button on the right.

Results

Use the Results tab to keep track of the data that the agent collects. Use this section to describe to the agent the specific information points that you want it to gather.

Add more information point gathering requests by clicking on the + Add result button.

Knowledge

Use the Knowledge tab to upload PDF and CSV files in order to give your agent context and knowledge materials in order to be able to answer questions or draw from existing knowledge in order to make decisions and carry out the task.

Click on + Add data source to upload what you need.

Setting your agent's triggers

Using the Triggers tab, you can tell your agent when to act. Your agent can be triggered on a set schedule, using the Embedded agent widget, creating a dedicated website for your agent, and more.

To learn more about triggers, please check out the Agent Factory automation triggers article.

Activity & insights

Activity & insights provide further insight into your agents' activities, tracked data, and results. 

Activity

Using the Activity tab, you cab view the Agent goal and view the Performance Metrics recorded for the agent's activity. 

Insights

Stay tuned! Insights are coming soon.

Testing your agent

Before publishing your agent, you can use the Test [agent's name] button on the top right corner to make sure it's working the way you'd like it to.  

Add the contact's name and phone number to test the agent. You can also click on Fill sample data to populate the inputs with sample data for testing.

When finished, click on Run test. You will then see a loading bar and the test runs. When finished, you will be able to view the test results under the What happened tab on the right.
You can then test again with the Re-run test button or make changes with the Improve agent button. 

Once you are satisfied with how your agent is running, click on Set Activation. You will be prompted to choose your agent's trigger.

Now your agent is ready to get to work!

FAQs

Is my data secure?

Agent Factory follows security protocols used across monday.com products. Data is encrypted at rest using AES-256 and in transit with TLS 1.3 to keep it secure during storage and transfer.

What are Agent Credits?

Agent Credits are the currency that powers your digital workforce. Each task an agent performs consumes credits, with usage based on task complexity (for example, phone calls are billed per minute).

What happens if I reach my credit limit?

If you exceed your available credits, agent activity is temporarily paused. You’ll be guided to add more credits or upgrade your plan to continue.

Is Agent Factory part of monday.com’s AI offering?

No. Agent Factory is a separate product. Other AI features like monday sidekick and AI blocks remain part of the core monday.com platform.

Can I connect my Agent Factory to my monday.com account?

Not at this time.

Does my existing monday.com contract cover Agent Factory?

No. Agent Factory is governed by its own terms, available under section 8 of the monday AI terms and conditions.

Which AI models does Agent Factory use?

Agent Factory uses Microsoft Azure OpenAI and Anthropic via AWS Bedrock, along with voice models from Vapi. monday.com regularly reviews providers to ensure quality and reliability.

Can I use my own AI models or API keys?

No. Agent Factory does not support custom AI models, external API keys, or third-party service accounts.

Will Agent Factory use my data to train AI models?

No. monday.com does not use customer data or content to train AI models and does not allow others to do so.

What are the limitations for using Agent Factory?

Agent Factory cannot be used for illegal or harmful activities, misrepresenting AI-generated outputs, automated decision-making in sensitive areas without human review, or providing professional advice without qualified oversight.

Human oversight is recommended for any use that could have legal, financial, or operational impact.

Pricing

Agent Factory uses a tiered payment model so that you can find the right plan and amount of credits for you regardless of the size of your business. You can always change the tier to meet your evolving needs as you scale your business. Please check out the Agent Factory pricing page to learn more.