Agents are powered by the same AI and organizational data that Coworker uses in chat — the difference is that Agents act independently, without you needing to be there. Rather than asking Coworker the same question every week, you build an Agent once, give it instructions and the right tools, and let it run.
When should I use an Agent?
The best candidates for Agents are tasks that are repetitive, time-consuming, or require pulling from multiple data sources. A useful rule of thumb: if you've done the same research or reporting task more than once, it's worth turning into an Agent.
Common use cases include:
Daily or weekly digests — Automatically summarize Slack activity, emails, or meeting notes and deliver a briefing each morning
Pipeline monitoring — Scan your CRM for stale deals and send a Slack alert when action is needed
Meeting follow-ups — After each meeting, automatically convert notes and action items into a follow-up email or Slack message
Status reports — Pull data from Jira, GitHub, or your project tools and generate a structured update on a recurring schedule
Customer health tracking — Monitor activity across accounts and surface at-risk customers before issues escalate
Inbox or document triage — Process incoming emails, invoices, or documents and route or summarize them automatically
Sales pipeline reports — Generate regular pipeline reports from CRM data and post them to a Slack channel
If you find yourself running the same prompt in Coworker more than once a week, that's a strong signal it should be an Agent.
How Agents work
An Agent has three core components:
Instructions — Plain-language directions that tell the Agent what to do, what data to look at, and how to format its output
Tools & data sources — The connected apps and data the Agent can access (e.g., Slack, HubSpot, Google Drive, Jira)
A trigger — What causes the Agent to run: a schedule, an @mention in chat or Slack, a manual run, or an external API call
When triggered, the Agent follows your instructions, accesses only the tools you've granted it, and delivers its output to wherever you've configured — the Coworker interface, a Slack channel, or your email.
Getting started with Agents
To create your first Agent:
Go to Agents in your Coworker workspace and click Create New Agent
Give your Agent a clear, descriptive name (e.g., "Weekly Pipeline Digest" or "Morning Slack Summary")
Write your instructions — describe the task in plain language, including what data to use, what to look for, and how to format the output
Select the tools and data sources your Agent needs access to
Choose a trigger — manual, scheduled, or on-demand via @mention
Save, run a test, and review the output. Adjust your instructions as needed.
Tip: Start with a single, focused task. Once your Agent is working reliably, you can expand its scope.