Want an agent to run automatically instead of launching it manually every time? Agent triggers — now set up on the dedicated Automations page — let you automate an existing agent so Pod checks your deals on a recurring schedule, evaluates a trigger condition, and delivers matching results to Agent Feed and, optionally, Slack.
How to access the Automations page 🤔
Open Pod.
Go to Settings.
Select Automations.
You can also land here from two shortcuts:
From Settings → Agents, click the ⚡ (Flash) icon on any agent card to jump straight into creating a new automation for that agent.
From Agent Feed, use Manage automations in the header.
If you still need to create the agent itself, start with Build and manage agents. To sharpen the prompt before you automate it, see the AI Agent Best Practices Guide.
Personal vs Org automations 👯
The Automations page opens on two tabs:
Personal Automations — automations you own, running on your deals.
Org-Wide Automations — workspace-wide automations. Only workspace admins can create or edit them; everyone else sees them read-only.
Use the status filter (All / Active / Inactive) to narrow the list, and the Manage Agents button to jump back to the Agents page if you need to change the underlying agent.
Create a new automation ⊕
On the Automations page, click Add Personal Automation or Add Org-Wide Automation depending on the active tab.
Select the agent you want to automate.
The Trigger Config drawer opens. Fill in the fields below.
Click Create Trigger.
If you are already viewing an agent in Settings → Agents, you can also click the ⚡ Flash icon on its card to start the same setup flow.
Configure the automation ⚙️
Recurring Interval
Choose how often Pod should check for matching deals:
Every 1 hour
Every 6 hours
Every 12 hours
Daily — set the time of day.
Weekly — set the day of week and the time.
Custom — set any number of hours or days between runs.
Pod shows the timezone it will use for the schedule (your browser's local timezone).
Trigger Condition
In Trigger Condition, describe in plain English when a deal should get this agent — the same way you'd explain it to a teammate.
You do not need special syntax, formulas, or CRM code. Write what you mean. Pod reads your wording and checks each deal using what Pod knows about that deal's recent activity (touchpoints, meetings, CRM updates, and similar signals).
What happens each time the automation runs:
The clock hits your interval (for example, every morning).
Pod looks at the records in scope and asks: “Does this match what the user described?”
Only the deals that match get the agent.
Run Mode (below) decides whether Pod runs again on the same deal every time it still matches, or only the first time.
Tips before you save:
Say what “worried” or “ready” looks like. Instead of “check this deal,” try “run when no one has touched the account in 10 days” or “run when the close date slipped twice.”
Add a time frame when it helps. Examples: “in the last two weeks,” “for at least 7 days,” “after the last customer meeting.”
Keep the condition and the agent aligned. This box = which deals should run. The agent's own instructions = what to do (summary, email draft, risk check). If those two don't match, you may get too many results or none.
Start broad, then narrow. After a few runs, open Agent Feed. If you're drowning in results, add detail. If nothing fires, loosen the wording.
Preview runs before you save
Click the Preview trigger runs icon (database-with-magnifying-glass) to see, for your current condition, how many agent runs Pod would plan and how many records are in scope. The banner reads something like:
This trigger will run 12 agents across 20 records in scope.
If the run density is high (more than 80% of records in scope), Pod shows the banner in orange so you can tighten the condition before you save.
Run Mode
Choose how often the agent should run for deals that match the trigger:
Always run — the agent runs on every matching deal each time the trigger checks.
Once per deal — the agent runs only the first time a deal matches.
Notifications
Turn Send to Slack on to receive a Slack DM after every successful automated run. This setting is on by default for new automations.
In practice, that's a one-line context header (agent, record, reason), the run's summary, and a deep link back to Pod where you can review the full result and approve or reject any pending actions.
Preconditions:
Personal automations — your workspace needs Slack connected, and you need to link your own Slack account in Settings → Integrations. See Manage integrations.
Org-Wide automations — the workspace admin can turn delivery on for the automation, but each recipient still needs their Slack account linked individually to receive the DM.
The drawer shows an inline alert when either precondition is missing so you know exactly what to fix.
Active or Inactive
New automations can be left Active so they start running right away. If you are still testing, leave the automation Inactive and enable it from the trigger card later.
When to choose Personal vs Org
Personal automations are best when you want an agent to monitor your own deals for your own workflow.
Org-Wide automations are best when the same automation should run across the workspace. Only workspace admins can create or edit them.
Manage an existing automation
Once an automation is live, return to the Automations page to manage it.
Click a trigger card to update its schedule, condition, Run Mode, or Slack delivery.
Use the switch on the trigger card to activate or deactivate it without opening the drawer.
Use the trigger menu to Configure, Edit agent, Duplicate, or Remove the automation.
The card shows the human-readable schedule and the last time the automation ran.
Use the status dropdown (All / Active / Inactive) to filter the list.
Where results appear
Results from automated runs appear in:
Agent Feed in the Pod web app and the Chrome Extension — see Use the Agent Feed for filtering, date grouping, and bulk actions.
Slack DM if you left Send to Slack on — click View in Pod → in the DM to jump back to the run.
Any pending actions proposed by the agent (email drafts, CRM updates, and similar) land alongside the run and wait for you to approve or reject them. See Manage agent actions.
When to use agent triggers 🤖
Agent triggers are most helpful when you want Pod to watch for recurring workflow moments without you having to remember to launch the agent manually.
Common examples include:
checking for stalled or inactive deals
watching for specific deal changes on a schedule
running the same analysis across the whole team with an Org-Wide automation
surfacing results in Agent Feed and Slack so you can review them later in one place
💡 Need help? Send us a message via the in-app chat or email us at [email protected].
🤝 Want to talk to someone? Book a session with one of our specialists.